Diary 2023-08-29
Parallel AI notes are coming together.
We're looking at the same Scrapbox, so that's what it would be.
I chopped it up into big pieces.
Yes - I'll red-link the phrase I think is right and let the "Generate red-link content".
Before page memory was introduced, "pages with robot marks were excluded from vector searches" in order to avoid reading one's own past notes. But is that still necessary today? State in which no connection can be found between parallel AI notes
Especially what I wrote in my AI Notebook Stimulus.
I took it off.
["🤖🔁Boundaries between humans and AI", "🤖🔁Humans and AI share goals", "🤖🔁Teaming with AI", " 🤖🔁Once chaos is brought about","🤖🔁2023-08-12 07:08"]
All I read are AI-generated notes.
Not always the case.
You designed it to "read your previous notes and develop your thoughts," but you're not limited to that if you're picking up something relevant.
Picking up chunks from multiple branches is fine, but picking up similar chunks from multiple pages in the main branch was a mistake.
The theory is that the implementation of creating a daily page for the main branch was a bad idea in the first place.
Talk about making books searchable.
A: Intellectual production techniques of engineers
B: PRAGMATISM
C: This Scrapbox
My brain envisioned finding a connection between A and C, but then I realized the quickest way to do it is B and C.
so I left it at that
The next step is to find a connection in A
This can be done by simply introducing the current script to A
The next step is to develop a bridge between A and C.
But if you don't want a daily up-to-date sync, you can just tell qdrant to put pickles from other projects into qdrant from the command line.
---
This page is auto-translated from /nishio/日記2023-08-29 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.