"Working out how to handle k..."


by Eric Bruylant Feb 12 2016

Working out how to handle karma on wikis is something I spent a bunch of brain cycles on a while ago. If you use only likes over the whole article it becomes impossibly unwieldy to fairly divide credit for content once you have a number of different contributors. Having things broken into small pages helps, but with like-only it still gives too little opportunity for rewards to people improving a page, and misses the chance for valuable engagement/positive reinforcement.

My favorite idea so far is rewarding user's edits with points. To avoid too much distracting subjective choice on the value of edits, having the recent changes review team (a necessary as anti-vandal feature) select from a pre-set collection of categories of edit when they confirm that an edit is not vandalism (providing a nice place for positive social feedback "thanks for your edit, your karma has increased by ~reviewer", in a more universal and naturally integrated way than https://en.wikipedia.org/wiki/Wikipedia:Notifications/Thanks). Examples of type of edit (each has a default karma value, maybe an acceptable range for reviewer to pick from): Typo fix, Add citation, Add section, Reorganize, etc. The relative default karma reward for each type would be adjusted in a way which the community feels is balanced in a domain-specific way. I've got a bunch of details and explanations of how it would handle potential problems if this seems like something you're interested in?


Alexei Andreev

Awesome! Yeah, I'd love to discuss this more when we'll start implementing this feature (later this year).

Another way to handle it is by having people vote on how big / important the edit was relative to the page's content. Let's say people decide the new edit added 10% value to the page. That means for the next likes, the editor should get 10% of that like and all previous editors should split the other 90% (recursively).

Eric Bruylant

That's another approach, but gets into annoying subjective judgements really easily. I'm confident it's harder to get volunteers to do "work out how much credit this person should get for this change, as a proportion of total value of the page" (better than just wordcount) than "you're already doing edit review to keep out vandals? also select from these edit types to categorize it" at scale with minimal oversight. Long term you could even do some machine learning or basic pattern recognition to work out what type of edit it is automatically most of the time so the reviewer just needs to confirm.

A hybrid approach where editing or creating high-value (by likes) pages gets more credit, by applying some scaling factor to the karma payoff, is plausible? Not thought that one through in detail though.

Nathan Rosquist

Just brainstorming: Some portion of the karma for an edit could be added retroactively by the % of page likes that happened after the edit. This would mean that (part of) karma is not a fixed thing, but sort of like dividends for the investment of time and energy in an edit.