16

Unlike questions and answers which may be upvoted or downvoted, there's no consequences for edit contributions - only rewards. There's no sense of responsibility for one's own actions.

All actions within the suggested edit model only reward with reputation or badges. There's no negative reaction other than temporary suspensions.

Given this model, it's not surprising many users proliferate trivial or useless edits that others approve.

Editors receive benefits such as:

  • +2 reputation gains on suggested edit approval
  • Badges: Editor Badge, Strunk & White, Copy Editor, Archaeologist, Tag Editor, Research Assistant

Reviewers receive:

  • Badges: Proofreader, Stewart, Reviewer, Custodian

With increased garbage-in from editors, more overhead is placed upon reviewers who also potentially gain via approvals, compromising the checks and balances of the system. With no accountability on either side, it's a vicious circle.

Reputation and badges encourage activities that are positive to the community, but with no consequences one can only positively grow their measure of trust within the community therefore gaining additional privileges.

If invalid or trivial edits were penalized by lowering reputation by 1, there would be an immediate impact on quality of edits. While less useful posts are downvoted, there's no downvote of less than useful edits.

Suggested edits should account for contributions that are not useful, subtracting reputation on rejected edits akin to a downvote.

  • 5
    I like the idea of suggested editor getting -1 for rejected edit. – CRABOLO Feb 27 '14 at 9:07
  • 3
    Once, my suggested edit was rejected by Community moderator automatically, because, while my edit was in review queue, a >2ker edited the post, without even going to review queue. I think in these cases, editors will be punished for nothing and also how are you going to punish those >2kers, who don't have to go to review queue??? – DroidDev Feb 27 '14 at 9:14
  • 4
    Rejections by Community could be excluded from this. >2k gets no reward so getting no punishment either would seem consistent. – Robert Longson Feb 27 '14 at 9:17
  • @RobertLongson shouldn't this be added to question? – DroidDev Feb 27 '14 at 9:21
  • 4
    @VishwasSharma, when that happens, it's because the person doing the review selected "Improve" to continue editing, and then unchecked the "this was helpful" checkbox, which causes the edit to be rejected automatically. It's still a rejection, even if it appeared to come from Community instead. Anyone trying to edit a post with a pending edit is forced to try to review the edit. There can be edit conflicts, but those are rare, and the proposed rep penalty should not apply when a conflicted edit gets squished. – Charles Feb 27 '14 at 9:33
  • 4
    Maybe review bans should be accompanied by reversal of reputation gain from suggested edits. – devnull Feb 27 '14 at 9:38
  • 3
    @Charles there are other race conditions that result in a rejected edit by Commnunity, it is not only cases where an edit was improved. – psubsee2003 Feb 27 '14 at 9:41
  • There's also anonymous users; however, those users have less incentive to gain reputation and badges. – Jason Sturges Feb 27 '14 at 9:44
  • 2
  • Why? Would you also put in a penalty for a bad edit that was done by someone who has the edit privilege? What about flagging, close votes, and other actions that gain badge rewards? – Joe W Feb 27 '14 at 17:00
10

As much as I like the idea of suggested editors being held accountable for their bad edits, there are too many problems in the existing system to make them the target of the "punishment" as this time.

  • Reviewers are more to blame for bad edit approvals. There is a reason edits made by users less than 2K rep (1K on beta sites) are suggested edits, it is because those users have not demonstrated a enough of a familiarity with the site to be trusted with editing posts on their own. We expect them to make some bad edits, otherwise there wouldn't be the need for a review.
  • There is no obvious feedback to the editor that their edits are getting rejected up until they get a review ban, and even then, the only feedback is a grayed out edit button. You need Decision on rejected edits should be displayed as a notification to the editor to get implemented first, then we can start educating editors more.
  • The edit reviewer situation is abysmal at best. When you have so many bad edits getting approved, you have a situation where the positive feedback (via badges and rep) is encouraging the editor to make more bad edits, and teaching them that the poor quality edits that have been making are acceptable. If you start deducting rep for making the same type of edit that they've gotten +2 on 10 times before, you are only going to confuse them.

So if we want to start "punishing" the editors, we need to improve the editing experience overall so it is much educational and much more consistent. When reviewers are consistently rejecting poor quality edits and when the editors are getting obvious and immediate feedback to their rejected edits, it is then time to discuss how to hold edits accountable.

So the key is improving the reviewers, so we can get the consistency in the reviews that is necessary, and that has proven difficult without having someone manually reviewing the reviewers.

  • 3
    Yes, that. We don't have a problem with too many suggested edits, or even with too many accepted suggested edits. The problem is with the lack of correlation between suggested edits that should be accepted and suggested edits that are accepted. I'd hate for someone who wanted to fix a typo in an answer that they'd tried to run to be discouraged not just by having their edit rejected but also losing reputation, adding injury to insult. Furthermore, this is pretty much an SO-only problem, which shows that it's a problem of culture more than a problem of incentives. – Gilles 'SO- stop being evil' Feb 27 '14 at 12:00
0

I like the general idea of punishing bad(robo) reviewers, people who accept bad edits and people who participate just to earn a badge causing more harm than help...

I like the idea of punishing those bad ones but only in cases where the punishment is correct. There are many cases where a user with good intentions could possibly be harmed by the system. The devs would need to test such system throughout to make sure nobody is punished for actually helping.

Vicious circle....
I think this game would have some negative impact on some/ if not most/ users. I wouldn't want SO to become a hazardous place for some users.

Pretend you're playing a one-armed bandit. You lost. You play again. You loose again... You get a little aggravated and want to get your money/coins back.

Now, realizing that there are a lot of people addicted to Stack Overflow game, I can see users slowly falling into ludomania... Increased traffic in the Suggested Edits queue wouldn't necessarily turn out helpful. It would possibly increase the amount of complains on Meta, chats, mods, emails etc.

The more are reduced -1 for accepting a terrible edit, the more are going to try to suggest an edit (or some other not necessarily proper/accepted by Stack Overflow action [1]) to gain the rep back.

While I agree that bad reviewers should get punished for approving bad reviews I am not entirely sure your proposed solution is the right one to implement. Thanks for reading.

[1] You don't want to see people going around other peoples profiles and downvoting their questions(cause you don't get -1 for it)/answers as a payback for their own stupidly lost -1.

  • Not sure I entirely agree with this. Applying the same logic, if users continually submit bad questions that are downvoted do we want more of those bad questions? They get question / answer banned which prevents further action. If useful contributions were rewarded, we'd get more useful contributions. – Jason Sturges Feb 27 '14 at 16:42
  • @JasonSturges useful contributions are already rewarded. I am sorry if I am missing your point though. – user221081 Feb 27 '14 at 16:52
0

I completely agree, it would be really useful to incentivate people to contribute in a better way to improve questions and answer, keeping in mind that a good edit suggest will be rewarded and a bad will result in a reputation loose.

  • Um, did you read the question? There is already an incentive (via rep and badges) to edit posts. The question is talking about deducting rep for making bad edits. – psubsee2003 Feb 27 '14 at 12:05
  • @psubsee2003: reading my answer again I see that it wasn't clear, I updated it. Thanks for suggesting. – Revious Feb 27 '14 at 13:02

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .