Prioritize what you use.
I had several ideas on how to proceed with the project, and I had a hard time deciding which one to go with, so I asked GPT4. As a result, an axis of conflict was discovered that I was not aware of.
Some of the options were ones that I considered "likely" to have the users I was seeking, but I myself was not a user.
One option, on the other hand, I was the user and beneficiary. I have clear needs of my own, not the "likely" needs of the former.
The problem is that there are two types of problems: "I don't know if there is a user, but I am sure I can achieve it," and "I have a user, but I am not sure if I can achieve it.
Then you should prioritize what you use.
relevance
Use it yourself and keep improving it.
Don't publish what you won't use.
Use your own deliverables.
from /omoikane/chat page 5.
2023/8/11
nishio.icon
Proposal for how to proceed after this
A: Write an article on how to install Omoikane Vecter Search
Anyone can do something like this in their own Scrapbox: Cases where vector searches were useful.
B: First, create an Omoikane Question Box
Omoikane Vecter Search with a question-box-like question-and-answer UI
What you get
Are there users who prefer it to mere search?
Someone wants to ask questions in nishio's Scrapbox (not sure if they really want to).
Someone may want to use their Scrapbox as a question box (unknown).
The process of virtual character co-management from Mr. Qualia's Question Box.
C: More trial and error with the Bot that writes to Scrapbox
Bot writes a note in Scrapbox and reads the note on the next run to become a "virtual character that keeps its memory".
This will be a unique Scrapbox experience!
gpt.icon
Option A focuses on education and awareness, B on community and dialogue, and C on technological exploration and innovation.
A: Useful when there is a lack of good teaching materials or when the emphasis is on disseminating vector searches
B: Useful if you want to enhance user interaction or are seeking new means of communication
C: Useful if you are looking for a technical challenge or want to provide a new experience utilizing Scrapbox
I see. Okay.nishio.icon
A and B think that the user they seek is "likely" to be there, but I myself am not a user.
A has been implemented and used long ago, so doing A doesn't increase utility.
I honestly don't think I'd use B because I think "search, then cherry-pick, put together in Scrapbox, then ask ChatGPT" has more utility.
The utility for me was that "someone asking a question" would allow Ms. Qualia to grow on her own, or provide an opportunity to develop her thinking.
However, Preparation for Omoikane Study Group found that asking ChatGPT a question can increase the linguistic fluency without the uncertain expectation that a human being will do it.
C is the beneficiary himself.
I've been hesitant because of the uncertainty, but until we see a concrete need for A and B, C should be our priority.
---
This page is auto-translated from /nishio/自分が使うものを優先せよ 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.