LENCHI_Day7
2024-05-22 Organize in Kozaneba
Kozaneba in [LENCHI_NEW: Giving a talk
https://gyazo.com/88ef1afdc46583c8d70994c8638025cf
This is just something that popped into my head when the need for a lecture arose.
The AI assistant just sees it and repeats almost the same thing, no new ideas.
However, "I think it's very good to talk about ~~!" It's good that there is a kind of encouagement going on.
I especially recommend it for those who have a hard time feeling confident about themselves and write it off.
The content has not changed significantly, but sometimes the same thing is written in a different way, and some of the expressions are "better" that way
For example, I didn't use the phrase "evolution of intellectual production methodologies", but Claude did.
This came out of my head, so it's just a reproduction of the structure I already have.
Adding an unexpected jolt in [LENCHI_Day6
https://gyazo.com/8cbe92095c29de527bc2f25a223a7c9d
The derailment story has come up many times.
https://gyazo.com/d47896b941af67d9bad005a1281d3f06
Trade-off in minutes whether to include the derailment story itself in the lecture or not.
succession
https://gyazo.com/e13892d6c7c42c1b871db865adf6b595
Lots of things in the tree that were not in the original tree.
Whether or not this is connected and whether or not it fits in time are two different things.
It probably won't fit, so we'll have to cut somewhere.
But I still have a few more things to add, so I'll get them all in first.
add (e.g. annex)
https://gyazo.com/a2429fb4119b4d8e04506ba5a88a8d58
merge
https://gyazo.com/2ee7e016b62f4ab1a06b95c6b5423ddf
add (e.g. annex)
https://gyazo.com/82ff610b8fbdfdbc81231885cb88edd8
merge
https://gyazo.com/36ca1b81a90ed22f2c61614fb2202105
https://gyazo.com/8a6e7af5d24bd2e9db5a018c22148119
The bottom right corner is getting bigger and more interesting, and the weight of the "stories that happened after the writing" that was originally planned may be going down.
But, well, my AI assistant says that if you suddenly talk about the lower right, the audience won't be able to follow you, so you need to talk about concrete experiences to build a foundation of understanding and then abstract step by step, and I think that's really true.
Fill in the X's with [I had LENCHI_LLM comment from a 2024 perspective.
https://gyazo.com/44feb6ee1f82ecc1c894b0d9479b61df
I put it down and wanted to rearrange it.
https://gyazo.com/f221b2ab575fd7ab4a2934d2b796b506
Did I mention that the summary is personal?
Also place human comments.
https://gyazo.com/6b7cd57c3b2d7672a089fb5aee5a1110
big picture
https://gyazo.com/af55ee623498d8e1122d96411ad730ce
I wonder if this can be image-recognized.
At this resolution, even GPT-4o doesn't seem to be able to tell much.
What about this one?
https://gyazo.com/095967d83181c64d943506fdb5c1fca3
gpt.icon(GPT-4o)
Specific sections of this mind map will be explained.
### Perspectives on 2024
- **IMPORTANT IMPLICATIONS**: a summary of key implications for the year 2024. Future directions and goals are presented here.
- **Exercising Creativity as a Human**: The importance of creativity and ways to exercise it are discussed.
### Understanding
- **Learning Cycles**: shows that the learning process is cyclical. Steps for deepening understanding are detailed.
- **Build Your Own Model**: describes building a model to better understand yourself.
- **Form your own ideas**: The importance of having your own ideas and unique perspectives is emphasized.
### Ability to structure and apply knowledge
- **Knowledge Structuring**: describes how to effectively organize and structure knowledge. Specific tools and techniques are included.
- **Applicability**: describes the ability to apply knowledge to solve real-world problems and generate new ideas.
### Willingness
- **Human Creativity Matters**: The value of creativity and the importance of motivation to exercise it are emphasized.
- **Ability to determine what one should learn**: The ability to determine what one should learn is important.
### Other elements
- **Cataclysmic Change in the World**: The world is changing dramatically, and the ideas and actions to respond to these changes are discussed.
- **Read Efficiently**: Specific techniques are presented on how to read and interpret information efficiently.
- **PURPOSE OF READING AND ITS OBJECTIVES**: This section demonstrates the importance of identifying and acting upon the purpose of reading and gathering information.
This mind map provides many perspectives on how to organize knowledge, deepen understanding, exercise creativity, and stay motivated. Please let us know if there are areas you would like to learn more about.
I don't see any advantage compared to seeing a human image.nishio.icon
-----
I have Claude read a short article from the past.
nishio.icon
Article written one year ago
claude.icon
I think this one year old article is very illuminating about reading and structuring knowledge using LLM.
First, it is noted that there is an aspect of the act of "reading" that involves not only skimming, but also deeply understanding and building a network of knowledge. The author has acquired a network of knowledge, called an "emulator," by taking the time to read for a limited number of authors.
By publishing this process in Scrapbox, they stated, they were able to get feedback from others and more effectively build a network of knowledge. However, in the case of short and difficult texts, the fragments are too small to be suitable for Scrapbox, which is a form of Kozaneba reading.
And the potential for further progress has been suggested by combining LLM: the hope has been expressed that LLM could be used to discover connections between fragments and to structure knowledge more effectively by having students ask questions about what they have written.
I think this article offers a very important perspective on the possibilities for intellectual production using LLM. It is thought-provoking in that it rethinks the act of reading from the perspective of building a network of knowledge, and points to the role of Scrapbox in this process and the new possibilities of combining LLM.
In particular, the idea of using LLM to discover connections between fragments and to respond to questions about what you have written may contribute greatly to the efficiency and depth of intellectual production. LLMs can complement the human thought process, and may even help you to gain new insights and ideas.
The content of this article is well worth addressing in a lecture. The subject of reading and structuring knowledge is of universal importance to audiences interested in intellectual production, and if you can show specifically how the process can be changed by using LLM, it should have a significant impact.
The author's own Scrapbox and Kozaneba reading practices will also be presented in the context of intellectual production using LLM to provide practical tips to the audience.
Please expand on the content of this article in your talk and discuss with your audience the new ways of reading and knowledge building that LLM will bring about. I am convinced that this will be an important step forward in envisioning the future of intellectual production.
nishio.icon
claude.icon
(Added to [There is more than one way to summarize.)
nishio.icon
claude.icon
Yes, let's use Kozaneba to organize and map the conversation so far.
First, keywords are extracted and grouped together.
LLM and changes in intellectual production
- Development of LLM represented by ChatGPT
- Changes in the nature of intellectual production
- Collaboration with AI Assistant
- The future is speckled
2. the value of "The Engineer's Art of Intellectual Production
- Long-term value content
- Importance of human creativity while harnessing the power of machines
- Implications from a 2024 Perspective
- Importance of knowledge that is difficult to change
3. change and importance of learning
- Transition of Information Access
- Contemporary Significance of the Learning Cycle
- Ability to identify subjects to study on their own
- Changes in the act of reading
4. evolution and utilization of tools
- Tools such as Scrapbox, GitHub, Kozaneba, etc.
- Past and present production tools
- Personal Knowledge Management Tools
- Learning from the history of tool changes
5. human creativity and motivation
- The Role of Humans in the Development and Automation of AI
- The process of generating and nurturing ideas
- Attitude of learning by making
6. intellectual production practices utilizing LLM
- Creating and utilizing custom GPTs
- Understanding and dealing with LLM constraints
- Balance between divergent and convergent thinking
- Possibility of LLM to maintain and converge objectives and context
7. devising of lectures
- Start with concrete examples and gradually increase the level of abstraction
- Use of metaphors and analogies
- Interaction with the audience
- Clarification of the overall structure
Placing these groups and drawing lines to show their relationships results in the following map.
code::
LLM and changes in intellectual production
|
|
Evolution and Use of Tools ---- The Value of "The Engineer's Art of Intellectual Production"
| |
| |
Practicing Intellectual Production with LLM ---- The Changing Nature and Importance of Learning
|
|
Human creativity and motivation
|
|
Lecture Devices
This map shows that while the development of LLM has brought about significant changes in intellectual production, the ideas advocated by "The Engineer's Art of Intellectual Production" continue to have value today. And the message emerges that it is important to utilize LLM while taking into account the evolution of tools, and to unleash human creativity.
The way we learn is also changing, and the ability to identify the objects of our own learning is becoming increasingly important; we will need to practice intellectual production while balancing divergent and convergent thinking through the use of LLM.
And in order to effectively convey these contents in a lecture, it is considered effective to interact with the audience by explaining step-by-step with concrete examples and utilizing metaphors.
The above is a map of the content organized using Kozaneba and the important points that can be read from it. Using this map as a guide, you can structure your presentation and aim for content that is easy to understand and thought-provoking for your audience.
(I honestly didn't expect this kind of output nishio.icon)
From my crude instruction "Kozaneba this time," it seems that he did not understand that the text I gave him was "a copy of the text resulting from Kozaneba by a human being" (I think it is natural that he did not understand), and that he tried to do Kozaneba in his own way, which he had grasped from our conversation so far.
They recognize that Kozaneba is a map that is created by extracting keywords, creating groups, placing them, and drawing lines.
You've got it about right.
nishio.iconOkay, so this is the Day 7 conversation so far: look back at what you did on Day 7. Also, think about what you would need to pass on if you were to let a new LLM continue and list the memories you would need to maintain throughout this project.
claude.iconDay 7 recap:.
1. the content of the lecture for the 5th printing commemoration of "The Intellectual Production of Engineers" on 6/6 was discussed.
2. the implications of the publication of a fifth printing and the impact of the development of the LLM, as represented by ChatGPT, on intellectual production were discussed.
3. changes that have occurred since the writing of the book include the use of Scrapbox, the practice of the KJ method in Kozaneba, and the emergence of LLM.
4. findings and issues from the dialogue with the custom GPT were summarized, pointing out the importance of balancing divergent and convergent thinking.
5. the changes in the act of "reading" and the importance of structuring and applying knowledge skills were mentioned.
6. preparation of lectures itself was a practice of "making and learning," suggesting the possibility of gaining new knowledge.
7. a deeper discussion was held on the relationship between the KJ method and LLM, and the usefulness of fractal summaries.
8. Suggested innovations in the lectures included the use of concrete examples, step-by-step abstraction, the use of metaphors, interaction, and clarification of the overall structure.
9. keywords of the lectures were organized and mapped using Kozaneba.
Information that needs to be transferred to the new LLM:.
1. purpose and outline: To share with the audience the possibilities and methodology of intellectual production using LLM as a commemorative lecture for the 5th printing of "The Engineer's Art of Intellectual Production".
2. main topics of the lecture: the development of LLM and its impact on intellectual production, the contemporary significance of "The Engineer's Art of Intellectual Production," changes in learning, evolution of tools, human creativity, and intellectual production practices using LLM.
3. devising the presentation: use of concrete examples, step-by-step abstraction, use of metaphors, interaction, and clarification of the overall structure.
4. progress to date: examples of the use of custom GPTs, the balance between divergent and convergent thinking, the relationship between the KJ method and LLM, and the usefulness of fractal summaries.
5. the author's awareness of the problem and direction of inquiry: to practically explore the possibilities of intellectual production using LLM and to present a new model of intellectual production.
Memories to be maintained throughout this project:.
Title: Commemorative lecture for the 5th printing of "The Intellectual Production of Engineers
2. purpose of the presentation: to share with the audience the possibilities and methodologies of intellectual production using LLM
3. main topics of the lecture: the development of LLM and its impact on intellectual production, the contemporary significance of "The Engineer's Art of Intellectual Production," changes in learning, evolution of tools, human creativity, and intellectual production practices using LLM
4. presentation devices: concrete examples, step-by-step abstraction, use of metaphors, interaction, clarification of overall structure
5. details of the process of working with LLM: dialogue with custom GPT, dialogue with Claude, process of preparing presentation materials, etc.
6. key concepts and issues: balance between divergent and convergent thinking, relationship between KJ method and LLM, usefulness of fractal summarization, etc.
7. the author's awareness of the problem and direction of inquiry: to practically pursue the possibilities of intellectual production using LLM and to present a new model of intellectual production
---
Here.
https://gyazo.com/7e4cade6cfbeadb0beaacdda3e84e1a1
Put it here.
https://gyazo.com/31abc4d0116f9120946842e67014e9d5
https://gyazo.com/5906f4b36d5b4cd2dbb2075963ad9895
https://gyazo.com/6086dc134e542cba761165b5788db39d
---
This page is auto-translated from /nishio/LENCHI_Day7 using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I'm very happy to spread my thought to non-Japanese readers.