data: people disagreeing on the Internet
Nov. 19th, 2006 10:36 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
guslMy research project, "learning argumentative structures", has generated lots of nice ideas, but not much in the way of concrete progress. It is time to get some data.
I want to see lots of instances of people disagreeing, and justifying their position (with real arguments: we are not interested in ad-hominem, but fallacious and unsupported arguments are ok), so I can make argument maps of it. I'm looking for blogs, mailing list archives, chat logs, etc in which people have discussions.
Desired properties of the data / medium:
(1) people use quotes, or somehow indicate precisely what sentence they disagree with.
(2) quotes are short, responses are short.
(3) redundancy: the same arguments and rebuttals appear many times using different words (linking, trackback all help here)
Some candidate source types:
* blogs: so-so about (1), so-so about (2). Good about (3), thanks to extensive linking and trackback.
* mailing lists: so-so about (1).
* chats: very good about (1) and (2), bad about (3).
I'm leaning towards making experiments with the chat medium, that somehow induce the subjects to talk about the same arguments. Hopefully, someone has done this for me already.
I want to see lots of instances of people disagreeing, and justifying their position (with real arguments: we are not interested in ad-hominem, but fallacious and unsupported arguments are ok), so I can make argument maps of it. I'm looking for blogs, mailing list archives, chat logs, etc in which people have discussions.
Desired properties of the data / medium:
(1) people use quotes, or somehow indicate precisely what sentence they disagree with.
(2) quotes are short, responses are short.
(3) redundancy: the same arguments and rebuttals appear many times using different words (linking, trackback all help here)
Some candidate source types:
* blogs: so-so about (1), so-so about (2). Good about (3), thanks to extensive linking and trackback.
* mailing lists: so-so about (1).
* chats: very good about (1) and (2), bad about (3).
I'm leaning towards making experiments with the chat medium, that somehow induce the subjects to talk about the same arguments. Hopefully, someone has done this for me already.