Conversation Summary 2023-04-28
An experiment in which GPT4 was asked to organize the traffic of discussions on Twitter and also to suggest next actions to make the discussions more productive.
prompt
The following data is a conversation between nishio and hrjn on Twitter. It contains a difference in interpretation of words, and the same word is used with different meanings. Summarize the conversation in bullet points, focusing on the similarities and differences between the two thinkers.
result
Common points:
- Interested in Decentralized Massmedia.
- We believe that new media are effective.
Differences:
- nishio believes that diversity of information is more important than the truth of fact-checking, but hrjn emphasizes the importance of fact-based reporting.
- nishio is not interested in defining the press, but hrjn considers the difference between the press and new media.
one more thing...
Suggest actions that nishio should take to develop the discussion.
Actions to be taken by nishio:
- Narrow the focus of the discussion: Because the discussion is too broad, it is important that nishio focus on specific topics and questions to establish a common understanding with hrjn.
- Provide examples: to support the claim that Decentralized Massmedia actually works, it is effective to cite existing examples and studies to demonstrate its effectiveness and benefits.
- Organizing and summarizing opinions: As the discussion progresses, it is important to summarize and organize as appropriate to make each other's opinions clear and easy for others to understand.
- Asking questions: To encourage further consideration of hrjn's opinions and questions, you can develop the discussion by asking questions.
- Respect each other's position: It is natural to have different opinions and viewpoints, so respecting each other's position and continuing dialogue will help develop a common understanding.
raw data
nishio: Most Japanese are not that opposed to government IDs, so their reaction to decentrized IDs is a lack of real feeling, like "Oh, I understand the logic behind the need for it". I think what many Japanese will feel more is decentrized massmedia, since they thought Twitter would be an alternative, but it wasn't.
@hrjn: I think this is true, but I'm not sure if the facts can be identified by crowd knowledge. In many cases, we are judging facts based on the credibility of the person explaining the facts, and in few cases, we are actually looking at the facts.
nishio I'm puzzled when I get responses saying "it can't be A" on a subject I think is not A.... How can it be "fact identifiable by crowd knowledge"? Why did you bring that up?
hrjn this?
nishio: I imagine that anyone can throw in a seed fact, and as it gradually spreads from a small number of people, it will be refined through the process of fact-checking, such as the ratio of approval to disapproval.
nishio If you interpreted this as "identifying facts", I never said that...
hrjn In order for "the process of spreading (omitted) to be generated by fact-checking and so on," the facts have to be checkable, which means that the facts It must mean that they can be identified.
What's a fact check that doesn't identify facts?
nishio Hmmm? I'm not sure what you mean, but what do you mean by "specific" in that case?
hrjn On the other hand, I'm not sure what you're trying to fact-check, but if you're saying, "The genuineness of the fact-checking is not important, but the opinions of various people If you're talking about "it doesn't matter how true the fact check is, it's helpful if it's collected", then I don't feel particularly uncomfortable.
nishio Ah, I see. Since the basic premise is this, "is it true?" is not important.
All data is a lie All data is a lie. We use data knowing that it is a lie. And from that, we can determine the more correct truth. That is the path to judgment. But how can the truth be determined from a lie? From a different point of view, every data has some shadow of truth in it. (KJ method, "Let the chaos speak for itself," p. 71)
hrjn I think it may be true that when various information is gathered, it is possible to somehow infer something plausible from it, but why it is so cannot be explained or structured in a rational way. I can't explain it rationally, and it can't be structured.
For example, it's a mystery to me if this was refined in the recent past.
nishio I've never seen that tweet before, but here's your "why yes", scrapbox.io Cumulative effect of ideas I think the concept expressed by the diagram of the interference effect of ideas in Jiro Kawakita's "Idea Method" is a very good one, but the diagram itself is not easy to understand. So I drew a new diagram.
hrjn Isn't the information handled completely different between a story of an idea and a media report that deals mainly with what seems to be facts?
"Should same-sex marriage be legally possible?" may have more than yes/no answers, so new media mediating diverse ideas may be effective, but it is not the same as reporting that deals mainly with what seems to be facts.
nishio Not at all -> Yes New media enabled -> Yes
Both are what I was trying to say, and I'm not interested in whether or not there's an argument about whether or not it's called reporting (and I wasn't talking about reporting in the first place).
nishio You should read it again, because it's really interesting how you "read what I didn't say and disagree".
nishio Oh, okay, I get it. So you saw this and interpreted "new things that have been replaced" as "reporting"? I think that interpretation is certainly possible.
nishio: I wonder if we can replace the media that does not report what it should do, and only reports what it shouldn't and what it doesn't give a shit about, with a combination of Polis and Mastodon or something...
hrjn Yes, that's right. I started out writing about replacing media coverage, so that's why I read de centerized mass media.
So, I've often wondered about similar things, but in the end, I'm talking about the fact that it's generally difficult to rely on the crowd for the certainty of information, so I feel like it's no different than mass media in the end.
This page is auto-translated from /nishio/会話の要約2023-04-28 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.