- Listening Chat System Experiment Log 20190610
We rewrote a lot of stuff over the weekend.
What kind of Saturdays and Sundays are those?
This Saturday and Sunday. It means a day off.
What else is there to say about that Saturday and Sunday?
I didn’t have to go out, so I was able to stretch out and write the program.
What type of program is that?
This is the chat program
What kind of chat is that?
A computer reads what a human has said, finds key words, and asks probing questions.
What type of computer is that?
Now it is running on Heroku.
What kind of Heroku is that Heroku?
I used them because they hosted the chat program for free!
What kind of hosting is that?
Free to run servers written in Python
What kind of gratuity is this gratuity?
I’ll put in a mechanism in the near future to indicate that I’d like you to dig deeper into the topic about the chat system, not this topic. As we dig deeper and deeper into the most recent keywords, we are getting further and further off topic.
What do you value?
Oh, good question. A system that listens to humans and asks good questions would allow a single person to deepen their thinking through dialogue. This creates value because it strengthens human intellectual productivity.
What is the relationship between that person and intellectual productivity?
What is the relationship between that person and productivity?
Humans consume resources from society just by existing, so humans need to produce and give back to society.
What is the relationship between that person and the resource?
Resources are needed to keep people alive.
What is the relationship between that resource and human beings?
NG This question is not good because I just replaced it.
What is the relationship between that question and NG?
NG Oh, that’s buggy, I thought I made it so that if you write NG, it doesn’t count as a normal input.
What is the relationship between that ng and the question?
- [Keyword and question selection is an internal volume caution. We fixed a few bugs in the process of porting a 4 year old system. I’ve made it so that you can have different states depending on the name of the other party and persist with Firestore. Ready for multiple people to use as chatbots.
This page is auto-translated from /nishio/人間と知的生産性の関係は？ 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.