Greetings everyone,
I'm a big Wikipedia fan and use it all the time to learn a little more about science or history or to find trivia about my favorite movies and TV shows. I haven't worked as a Wiki contributor, but I've heard a lot about the problems caused by multiple users editing highly controversial articles like the one about George W. Bush, and of course the implosion that happened with the L.A. Times "wikitorial" on the Iraq war.
What I've been working on is a collaborative algorithm for editing controversial content, without back-and-forth flame wars spiraling out of control, and without aspects that many in the Wikipedia community would prefer to avoid (e.g. without rating people by their professional "credentials", and without moderators who can permanently close disputes resolving them in favor of one side or the other, which I understand many Wiki users regard as a necessary evil even when they do have to be invoked). I'm testing it out by seeding it with some editorials that many people are sure to have strong feelings on (e.g. libertarianism, the Iraq war, and -- God help me -- Windows vs. Linux), and I'd like to see if some Wiki community members would be interested in signing up and trying it out.
The forum is at www.brainjammer.com and the "secret code" to sign up for a new account is "muskrat".
The way the forum works is simple: You can post a message, and other users can post replies to your message. When another user posts a reply, you have several choices: (a) Agree with their reply and modify your own post to incorporate their point. (This is also what you can do if the user has raised an objection but you can modify your argument to answer that objection, or if they have made a request for clarification and you are updating your post accordingly.) (b) Withdraw your own post in response to their reply, if you think that the flaw they have found in your argument is fatal. (c) Indicate that you disagree with the reply if it's a rebuttal. If this happens, then a "jury" of other users on the site will be selected at random to vote on who is right, in accordance with rules that are posted on the site.
So far, nothing very original. But the key rule that governs the rebuttals and the voting, is that a rebuttal to someone's post should focus on only *one* flaw. If someone posts a long and complex argument, and you post a rebuttal that highlights only one minor point in the argument that you think is wrong, but the original author refuses to concede the point and it goes to a "jury vote" and the jury votes in favor of your rebuttal, then the entire original post is "disqualified". The author can, however, re-post the essay without that flaw, if they think their argument holds up without it.
Why do it this way? My theory is that people will think (and vote) more honestly when focusing on one *highly specific* disputed point. For example, if one user writes a pro-Bush essay and a second user posts a "rebuttal" that is just an anti-Bush essay, then if you ask people to vote on who is correct, people will pretty much just vote their original prejudices. But if the first user writes in their essay that Bush voted in favor of a bill, and someone post a rebuttal citing a record showing that Bush actually voted against it, and the first user doesn't concede the point and it goes into a vote-off, you can be pretty sure that most users will vote in accordance with the factual record. Similarly, the theory is that people will vote more honestly not just on questions of fact but logic as well -- for example, if one user attacks Bush for "signing such-and-such into law", and a second user replies that Bush was on record as opposing such-and-such but was part of an omnib! us spending bill and that Bush had no choice but to sign it without shutting the government down, then if the first user refused to concede the point, I would expect most users to vote in favor of the second user's argument that the first user's statement was misleading.
My thinking is that if this works even for controversial writings like editorials for or against the war, it could help prevent vandalism and flame wars in other peer-edited systems like Wikipedia. And the structure remains completely democratic, allowing disputes to be decided by a jury of fellow users -- on the theory that users vote honestly when voting on highly specific questions.
So, if you have a minute, please come try it out. I've put a lot of work into this in the hopes that it could help projects like Wikipedia go even further, but I can't test it myself without a critical mass of users.
Thanks and I hope some of you are as excited about this idea as I am! :)
-Bennett
Bennett Haselton wrote:
(c) Indicate that you disagree with the reply if it's a rebuttal. If this happens, then a "jury" of other users on the site will be selected at random to vote on who is right,
One reason that voting is so often avoided in online communities is that it is unclear who is eligible to vote. While one-person-one-vote is a beautiful principle, it is not self-evident who can be counted as "a person". How do you convince your users that the voters aren't just sock puppets, or that they are of a mature enough age to make an informed decision?
Lars Aronsson wrote:
Bennett Haselton wrote:
(c) Indicate that you disagree with the reply if it's a rebuttal. If this happens, then a "jury" of other users on the site will be selected at random to vote on who is right,
One reason that voting is so often avoided in online communities is that it is unclear who is eligible to vote. While one-person-one-vote is a beautiful principle, it is not self-evident who can be counted as "a person". How do you convince your users that the voters aren't just sock puppets, or that they are of a mature enough age to make an informed decision?
Most people do not make informed decisions anyway IMHO. It is all about emotion. Someone can be a great admin never abuse his admin buttons but still be voted out of adminship because people do not like the person. Viceversa also happens.
Waerth
At 03:15 PM 7/12/2005 +0700, Walter van Kalken wrote:
Lars Aronsson wrote:
Bennett Haselton wrote:
(c) Indicate that you disagree with the reply if it's a rebuttal. If this happens, then a "jury" of other users on the site will be selected at random to vote on who is right,
One reason that voting is so often avoided in online communities is that it is unclear who is eligible to vote. While one-person-one-vote is a beautiful principle, it is not self-evident who can be counted as "a person". How do you convince your users that the voters aren't just sock puppets, or that they are of a mature enough age to make an informed decision?
Most people do not make informed decisions anyway IMHO. It is all about emotion. Someone can be a great admin never abuse his admin buttons but still be voted out of adminship because people do not like the person. Viceversa also happens.
Yes and this is exactly why the system only allows voting on highly specific questions of fact or logic and not just whether you "like" someone. The theory is that people will vote more honestly on specific questions one at a time, where there's less room for rationalization.
This theory may or may not end up being correct, but that's why I'm encouraging people to sign up and try it out.
-Bennett
bennett@peacefire.org http://www.peacefire.org (425) 497 9002
wikipedia-l@lists.wikimedia.org