• Preparation for Omoikane Study Group
  • I had ChatGPT create questions for me to generate material for my presentation.
    • I put the answer to that question in Scrapbox.
    • This was repeated to create the material.
    • When I had enough, I thought of a composition in Kozaneba.
    • I made it using only the parts I used while looking at Kozaneba.
    • At this time, of course, “questions and answers that were written but not used” can be created.
    • It is beneficial that this pair be reused
  • This scrapbox is automatically chunked by Omoikane Embed and indexed by vector search.
  • If Scrapbox is automatically inscribed into a chunk and subject to a vector search, then the next time you try to write about a related topic, GPT can pick up that chunk in a vector search and reuse it.

from [/omoikane/chat page 4#64cc7a36aff09e0000cdb196](https://scrapbox.io/omoikane/chat page 4#64cc7a36aff09e0000cdb196)

Once this is done, the source of value moves from the lecture material to the pool of questions and answers in the foreground.

  • It didn’t seem like anything special.
    • Already in the days of Tadao Umesao, the source of the value of intellectual production was not the output book but the card-type database in the back.
  • It was just difficult to observe and distribute it.
  • Right now, in an ongoing process, Nishio’s “card-type database” is observable in the form of Scrapbox.
    • Interaction with the reader is happening at the card stage.
  • A little further into the future, “Interaction with readers” will become “Interaction with AI”.

This page is auto-translated from /nishio/ChatGPTとScrapboxの効果的な連携とコンテンツの再利用 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.