There were references to a number of documents that were somehow loosely related, so I summarized them all in GPT4, summarized, summarized, summarized, and then had the page titles made up as I went along.
- Toward Efficient information sharing and [Verbalization of tacit knowledge
- The person being questioned leaves a note. This will increase efficiency, and the difficulty of verbalizing tacit knowledge (knowledge management terminology) and the importance of documentation will be taken into account. The language model is expected to facilitate the expansion of documents from query analysis, while taking into account the difficulty of verbalizing tacit knowledge (knowledge management terminology) and the importance of documents.
query 質問
- The person being questioned keeps notes, which increases efficiency and accepts repeated questions.
- Tacit knowledge is difficult to be aware of and may be difficult to verbalize
- Documentation is important and could move in the direction of being expanded by the language model
- A future in which LLM summarizes readers’ queries and makes suggestions for document expansion is anticipated.
- Careful use of “documented” and “undocumented” functions, proper documentation facilitates maintenance
Raw log
0xtkgshn Which is the Scrapbox page for concepts like “If you ask them questions all the time, they’ll start writing, so ask them questions all the time”?
0xtkgshn @nishio I looked for Scrapbox so hard but couldn’t find it. 0xtkgshn Oh, maybe this? [/villagepump/the person being questioned leaves a note](https://scrapbox.io/villagepump/the person being questioned leaves a note). 0xtkgshn It’s a little different, I can’t remember what it is.
0xtkgshn”You should ask people the same thing over and over again.” “Don’t ask people the same thing over and over!” but that’s not true. It is more efficient for the person being asked to write a wiki, so the person asking should ask the same thing many times without taking notes. >> Verbalize what you don’t understand and let the other person answer. [/yuta25/You should ask people the same thing over and over again](https://scrapbox.io/yuta25/You should ask people the same thing over and over again)
0xtkgshn [/mrsekut-p/the person asked leaves note](https://scrapbox.io/mrsekut-p/the person asked leaves note)
0xtkgshn [/sta/ tacit knowledge is hard to be aware of](https://scrapbox.io/sta/ tacit knowledge is hard to be aware of)
0xtkgshn # so write
nishio: This pathway is important, a future where LLM summarizes the query the reader is throwing and suggests “you should write more about ~” or something! twitter.com/dmikurube/stat…
0xtkgshnI don’t understand what people who understand don’t understand
This page is auto-translated from /nishio/効率的な情報共有と暗黙知の言語化に向けて 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.