I’m new here and might not understand the present karma system completely or correctly. I like it, in certain ways, but I also know, from long internet history, that systems like this can be abused.
A well-known and acknowledged internet troll just openly threatened (on RationalWiki, where I’ve retired) to come here and harass me. I know what he does. I’m not concerned about argument from him, the karma system will handle that. However, he will also do these things, it can be predicted:
*He will look at all my past contributions and will down-vote them as much as allowed.
*He will register new accounts as needed. He’s highly skilled at this.
*He will look for any method of gaming the system, he will probe for vulnerabilities.
*He doesn’t care about the site purpose. He cares about winning a game.
The present karma system looks vulnerable to activity like this. I don’t see any clear sign that he’s been active so far. I reached a nadir of about −40, which is not surprising, I had raised certain issues that might be unpopular here. I modified my behavior, that’s a positive effect of the karma system. I’m at −2 right now.
His threat might be empty. However, these are the problems with the karma system that I see:
The voting is anonymous and there is no accountability. There is a suggestion that’s been made that downvoting should have a cost. It’s possible that all voting should have a cost. Otherwise we get voting (an action with consequences) with no personal responsibility (leading to a weighting toward people who really don’t care, but just respond, knee-jerk, possibly irrationally).
Voting systems ideally represent what happens in the brain. We have affective and aversive responses, and we do make decisions based on the overall weight. However, rational process, internally, can look at each response and value or devalue it, and the same happens in social processes with responsible actors. In the karma system, there is no way to look at what is producing up-votes and down-votes, and most votes are not accompanied by any comment at all.
Voting is presently three-valued, like Range 3 voting: i.e., values of −1, 0, +1. While this can be a great voting system (substantially better than binary), the total votes in each category, in real systems, can make a big difference in subsequent process. I.e., a net of −1 based on a single downvote, is a very different creature than the same net with 50 ups and 51 downs. The latter is probably of high interest! It would indicate a true divided community, as distinct from one that doesn’t care, it could indicate an area that needs more discussion. If it’s +50 and −53, it would indicate the same thing, the difference is in the noise, but now the karma system would inhibit the very discussion needed.
This leads to some immediate suggestions:
Report the votes in each category, not just the total.
Increase the resolution, i.e., say, allow double voting in each category of vote, and categorize these separately. (The system then becomes Range 5.)
Consider systems that make users more accountable. Perhaps report for each user how many upvotes they cast and how many downvotes. Or even make voting not anonymous. In real deliberative organizations, all opinion is public, and secret ballot is never used for issues, only for certain kinds of elections. An Objection to Consideration of the Question, for example, is subject to immediate, public vote. And if someone still wants to raise the issue, they know to whom to talk, individually. That is a device that increases social intelligence (distributed conversations).
Use a percentage rule for consequences of vote totals, rather than a fixed difference.
It is a general situation that internet process is at a primitive stage. Moving toward simulation of intelligent decision-making process could greatly improve the effectiveness of any society.
The karma system is a great step toward this, but appears vulnerable in certain ways.
I’m not sure it will do much good, but here is the post, and this is a permanent link to the discussion as it stands now. This was a goodbye post, to AD, one of the seemingly saner members of the RationalWiki community, an elected moderator. There is a link in my goodbye post back to AD’s comment in a discussion that included history, but that’s a lot more than I expect people here to be interested in. Suffice it to say that the user has a history of being exactly what he says he is, a highly effective troll. He says “professional.”
(To understand some of the discussion, “promote” on RatWiki means “remove sysop privileges” or sometimes “block.”)
I’m new here and might not understand the present karma system completely or correctly. I like it, in certain ways, but I also know, from long internet history, that systems like this can be abused.
A well-known and acknowledged internet troll just openly threatened (on RationalWiki, where I’ve retired) to come here and harass me. I know what he does. I’m not concerned about argument from him, the karma system will handle that. However, he will also do these things, it can be predicted:
*He will look at all my past contributions and will down-vote them as much as allowed.
*He will register new accounts as needed. He’s highly skilled at this.
*He will look for any method of gaming the system, he will probe for vulnerabilities.
*He doesn’t care about the site purpose. He cares about winning a game.
The present karma system looks vulnerable to activity like this. I don’t see any clear sign that he’s been active so far. I reached a nadir of about −40, which is not surprising, I had raised certain issues that might be unpopular here. I modified my behavior, that’s a positive effect of the karma system. I’m at −2 right now.
His threat might be empty. However, these are the problems with the karma system that I see:
The voting is anonymous and there is no accountability. There is a suggestion that’s been made that downvoting should have a cost. It’s possible that all voting should have a cost. Otherwise we get voting (an action with consequences) with no personal responsibility (leading to a weighting toward people who really don’t care, but just respond, knee-jerk, possibly irrationally).
Voting systems ideally represent what happens in the brain. We have affective and aversive responses, and we do make decisions based on the overall weight. However, rational process, internally, can look at each response and value or devalue it, and the same happens in social processes with responsible actors. In the karma system, there is no way to look at what is producing up-votes and down-votes, and most votes are not accompanied by any comment at all.
Voting is presently three-valued, like Range 3 voting: i.e., values of −1, 0, +1. While this can be a great voting system (substantially better than binary), the total votes in each category, in real systems, can make a big difference in subsequent process. I.e., a net of −1 based on a single downvote, is a very different creature than the same net with 50 ups and 51 downs. The latter is probably of high interest! It would indicate a true divided community, as distinct from one that doesn’t care, it could indicate an area that needs more discussion. If it’s +50 and −53, it would indicate the same thing, the difference is in the noise, but now the karma system would inhibit the very discussion needed.
This leads to some immediate suggestions:
Report the votes in each category, not just the total.
Increase the resolution, i.e., say, allow double voting in each category of vote, and categorize these separately. (The system then becomes Range 5.)
Consider systems that make users more accountable. Perhaps report for each user how many upvotes they cast and how many downvotes. Or even make voting not anonymous. In real deliberative organizations, all opinion is public, and secret ballot is never used for issues, only for certain kinds of elections. An Objection to Consideration of the Question, for example, is subject to immediate, public vote. And if someone still wants to raise the issue, they know to whom to talk, individually. That is a device that increases social intelligence (distributed conversations).
Use a percentage rule for consequences of vote totals, rather than a fixed difference.
It is a general situation that internet process is at a primitive stage. Moving toward simulation of intelligent decision-making process could greatly improve the effectiveness of any society.
The karma system is a great step toward this, but appears vulnerable in certain ways.
Could you give a link?
I’m not sure it will do much good, but here is the post, and this is a permanent link to the discussion as it stands now. This was a goodbye post, to AD, one of the seemingly saner members of the RationalWiki community, an elected moderator. There is a link in my goodbye post back to AD’s comment in a discussion that included history, but that’s a lot more than I expect people here to be interested in. Suffice it to say that the user has a history of being exactly what he says he is, a highly effective troll. He says “professional.”
(To understand some of the discussion, “promote” on RatWiki means “remove sysop privileges” or sometimes “block.”)