How do we know that the early stages are not going well? Perhaps if we compare the logs that are going well with those that are not, there is a big difference in the growth of the number of keywords extracted, and that is how we can determine this.
What would you like to see happen in this conversation?
I would like to think about the design of the early stages of the project.
What type of early stage is that early stage?
In the early stages of this dialogue, some people don’t get on track well and crash, and I want to help them get on track well.
Is there anything else you can tell us about that early stage?
The chat system asks in-depth questions in response to keywords, but in extreme cases, for example, there are people who enter sentences with no keywords at all. In such cases, of course, there are no keywords, so it is impossible to ask a proper question.
In the early stages of this dialogue, some people don’t get on track well and crash, I want to help them get on track well.
Where is this early stage?
Especially the first time you use it, I recommend looking at the log, but you start it up without looking at it and put “what do you want to happen?” and then you start up without looking at the log and put “Nothing” in the “What do you want to happen? Well, aside from such extreme cases, even people who have read the logs to some extent sometimes use NGKW too much and run out of keywords.
In the early stages of this dialogue, some people don’t get on track well and crash, I want to help them get on track well.
How do you know these early stages?
(Why such a focus on the early stages?)
As for how to know that the early stages are not going well, perhaps you can compare the logs that are going well with those that are not and see if there is a significant difference in the growth of the number of keywords extracted, and then you can determine that.
Where is the beginning of this process?
🙁
How do you know that the early stages are not going well? If you compare the logs that are going well with the logs that are not going well, there is probably a big difference in the growth of the number of keywords extracted, and that’s how you can tell.
What does this early stage look like?
It’s already there, but it’s like launching a rocket, you have to get a certain speed or you’ll crash.
What type of rocket is that?
Rocket to reach the moon
Is there anything else you can tell us about that rocket?
I can detect lack of speed, but I don’t know what to do when I detect it.
What type of speed is that?
Speed of keyword growth
Is there anything else you can tell us about its speed?
On the other hand, some people reduce it more and more.
What type of person is he/she?
I’m too rigid in my evaluation of keyword extraction, and if something is even a little bit wrong, I’ll NGKW more and more. I’m not in the business of creating training data for a keyword extraction engine, but using this chat system is supposed to help me reach awareness (month), but I’m acting counterproductive.
What type of chat system is that?
Take out what humans have not yet verbalized.
What is it that you value?
By using this chat system, I can verbalize what I have not been able to verbalize yet, notice my blind spots, etc., and I have a high probability of this happening, so I want other users to be able to use it well and experience the awareness.
Is there anything else about that chat system?
Hmmm, does it take practice to get used to using it?
By using this chat system, I can verbalize things that I have not been able to verbalize yet, or notice blind spots, or something like that, and I have a high probability of this happening, so I want other users to be able to use it well and experience the awareness.
What is this chat system like?
I am a counselor who can chat with him 24/7, but his Japanese is a bit awkward. If I had to point out every single mistake in her Japanese, the conversation would never get anywhere.
By using this chat system, I can verbalize things that I have not been able to verbalize yet, or notice blind spots, or something like that, and I have a high probability of this happening, so I want other users to be able to use it well and experience the awareness.
Where does this chat system come from?
This page is auto-translated from /nishio/序盤のデザイン 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.
Especially the first time you use it, I recommend looking at the log, but you start it up without looking at it and put “what do you want to happen?” and then you start up without looking at the log and put “Nothing” in the “What do you want to happen? Well, aside from such extreme cases, even people who have read the logs to some extent sometimes use NGKW too much and run out of keywords.
Where is the beginning of this process?