You are right, ELO doesn’t quite fit.
I guess what I was thinking was just a generally weighted system and ELO was the one that came to mind that used that concept.
How I see it:
You have a baseline of points, say 1k.
@ 1k you get a 5x multiplier to points you get (to incentivise new users)
@ 2k - 2x multiplier
@ 3k - 1x multiplier
@ 4k - 0.5x multiplier
@ 5k - 0.1x multiplier
And you lose a set number of, say, 100 points per day, though you can never drop below 1k.
Additionally, to get the first 1k points:
You register - 100p
You go through the tutorial (showing guidelines, teaching how Slant systems work, etc.) - 900p
Where points show how good you are currently, badges would add lasting accomplishments.
You can hold 3k+ points for a week? Badge
4k+ points for a week? Badge
For a month? Badge
Get most "thank you"s? Badge
Most contributions within a category? Badge
Etc.
Edit: I should add that the baseline should probably be a lot higher and the weights to kick in after certain number of criteria - so that an everyday contributor would start feeling the effects of weights after, say, a month and someone who only occasionally adds content would not feel them at all. So the weights basically start becoming a thing when reaching power-user amounts of points.
1 Like
hrmm, this is more complicated of a system than I was envisioning but I do see the benefits of this approach. One issue I have is we’re effectively penalising people not constantly using the site. As @johnlbevan said:
As a founder I obviously like the idea of having users compelled to constantly use the site…but I can’t help but think there might be some backlash? Thoughts?
Also still thinking about:
In my v1 idea above I did give X points per edit, but maybe that is the wrong approach and we should do something along the lines of what @hethegreat suggested. Wikihow has a “patrol recent edits” feature where the mods can review and approve edits. Maybe we need something like that before we can include edits in the karma ranking.
For an initial push I think a point cap for edits would be satisfactory. Leaving it off would probably be best to keep points mostly pure.
Ok so we have a new concept. What if we used the point system I described above to calculate “Top contributors” for each option. So it’s less a race to acquire points, but more to go around and focus on a few options you care about.
The UI would be a little more badge based, and you could semi-compete foursquare style to be the main contributor for options/categories you care about. I think it’s a nice system as it gives people ownership over content they put effort into.
Profiles would highlight your best contributions etc and we might have leaderboards as well etc.
Thoughts on the concept?
3 Likes
I agree. Using badges that you can work towards is a great way to keep people contributing. Encouraging them to own a topic and keep it updated will certainly help the site to stay current and give users a sense of accomplishment.
Also, possibility to give out bounty (+100 points for example) for tackling certain questions.
Ok we’re finally getting around to building a v1 for this. We eventually want to do a proper badge system, but in the interim we’re going to build out a more simple karma system. It’s going to work on the concept of “validated” contributions. You get Xpts for adding something, then X + Y when we get a signal that it was a good contribution. In general, X < Y. Here is what the initial system will look like:
- Ask Question - 20 points - Another 10 points for every person who watches the question.
- Add Option - 40 points - Another 40 points if at least one other user votes for it. Another 20 points if it’s the first option in the question.
- Add Pro - 30 points - Another 20 points if at least one other user votes for it. Another 20 points if it’s the first pro for the option.
- Add a source - 10 points - Another 10 points if it has a quote. Another 10 points if it’s the first source for the option.
- Add the description to a Pro (that previously didn’t have one) - 10 points.
If the contribution you added is spam, when it’s removed the karma will also disappear. Thoughts?
3 Likes
We’re also going to use this data to have a weekly leaderboard on the sidebar (replacing featured users).
1 Like
Sounds good.
One question - with Pro (& Con), you get points for being the top Pro. That may change over time - do you lose the points when another pro bubbles up to first place / what happens if there’s a tie for first? Similarly for Sources.
It’s not the top rated pro that gets more love, but the very first one that was added. We want to encourage/reward the people who get the ball rolling and add the initial content.
3 Likes