Reputation for editing?

Question

The recent 'frank exchange of views' about downvotes suggests another line of investigation to me.

I would presume to attribute some goals to Jeff:

1. Users who put real effort into the site should earn reputation and achieve moderation status.
2. Users who game the system should not achieve moderation status, and asking a lot of toss-off questions is more or less gaming.
3. Innocents should not get caught in the crossfire and either (a) feel motivated to walk away or (b) feel motivated to flood Jeff with email complaining about harsh treatment.

Answering questions most assuredly counts under #1. I submit for consideration that careful editing also comes under that heading. However, there is no voting/reputation pipeline for this.

What if it were possible to vote edits up and down, with a suitable rep threshold? This would provide an additional way to take note of the real 'gardeners' of the community without having to hand out painful downvote raspberries to more or less innocent others.

Net possible result: more separation in reputation-space of the people who should get mod powers from those who don't without more injured email.

Other possible result: a reopening of flame wars about appropriate & inappropriate edits.

Clarification

Various answers and comments motivate the following:

I don't want reputation here. I'm not that faithful of an editor. I'm just following a trail of mental breadcrumbs from Mr. Atwood's posting about users who find downvotes really upsetting. This is presented as a possible alternative to give reputation where it seems to be wanted without downvoting people likely to squeak.

If the threshold for voting on edits is high, then the amount of three-ring circus thus produced might not be too awful.

Answer 1

Even an additional badge for editing has been viewed as too much incentive (and I'd reluctantly agree), causing more harm through poor edits than encouraging good ones. Rep would increase that problem tenfold.

As someone who frequently edits (compared to the norm for 2k+ users; at least it feels that way), I have to say that I don't mind not getting rep and it would be nice to get more badges—but I'm already editing without them, and will continue without extra incentive.

Answer 2

There's already a hidden incentive for editing: you have the power to edit other people's stuff! How many other sites on the internet grant that ability?

It's like pair programming, really.

Answer 3

I don't imagine very many edits would ever be voted on. Every so often, though, I see an edit that so dramatically improves a question that I wish I could reward it directly. – Bill the Lizard

In addition to leaving comments for editors, the poorly-named find-an-upvote-worthy-post-to-compensate-for-appreciated-non-votable-action principle does allow indirect rewards for a good or great edit.

Rolling back poor edits only takes 2k rep, and they can be flagged for attention. This seems sufficient—in terms of fixing that mistake if not in preventing more—so I won't address poor edits any further.

But for those great edits where a direct reward is desired, can you reward quality instead of volume without involving rep? Don't address any sort of punishment or disincentive, and allow only the equivalent of an upvote—think how "great comments" work.

Don't show the button without viewing the edit history, and thus the button would be on the revisions page. You'd have to go through a tiny bit of work to even recognize the editor's effort—that's a good thing.

Tie it into badges for the editor equivalent of the pundit badge: +10/+100/+1000 for bronze/silver/gold. Allow multiple gold awards, if you really want to recognize prolific editors, or cap at a silver badge (+10 bronze, +300-500 silver?) with multiple silver awards. Initially this seems hard to get, but remember we're counting votes-on-edits rather than counting edits, and pundit is hard to get too: only 60 on SO currently.

Though the parallels with the pundit badge are attractive, it seems that most comments either need to be snarky or controversial to hit +10 (which makes sense, easier to get people to vote emotionally), so I don't want to draw too close of a correlation. Importantly, I want nothing along the lines of "40 edits with +10 or higher score"—instead I'd like to recognize that most work lies in the long tail.

Answer 4

On other SE 2.0 sites, additional incentive to help keep the place clean might make more sense. I still have reservations about additional incentives (or possible deterrents) for editing for the same reasons that others have stated.

What I haven't seen anyone raise is, Stack Overflow is a community of programmers. If programmers see something incorrect, they (typically) naturally want to fix it. For the most part, I think we just fall into a natural sense of order when it comes to editing collaboratively.

This is interesting because what may work beautifully on Stack Overflow might not work so well on a site about various types of pets, model railroading, etc. I think as the betas progress, we will quickly see how much the 'programmer factor' plays into the success of SO, which favors being correct above everything else. Rewards and even sometimes fairness are and always have been a secondary, if not incidental consequence of using the system.

Answer 5

Currently, I read questions and answers (and write some), and then upvote and downvote as I see fit. To vote intelligently on edits, I'd have to examine all the edits, which is not going to be as easy or natural.

In short, I don't see how it would work well without complicated mechanics, which I definitely don't want.

Answer 6

I edit fairly extensively, and I trust my edits are mainly constructive. I'd like to see some recognition beyond the Strunk & White badge for the work done.

I don't know if a discretionary badge could be considered - maybe one that is triggered by someone completing another thousand edits (of other people's material). A notification could be sent to some team (maybe the moderators, maybe some other team) which says "this person could be eligible for the 'Senior Editor' badge; please check", with the team able to nix the award if the edits are seen as gaming the system rather than being constructive. There'd need to be some tooling to help them review things quickly - and that may not be so easy. Id suggest that the normal presumption should be "yes, the user earned it", but maybe leave the team a week in which to review any specific award. And yes, I think it would be nice to recognize ongoing effort, so the badge should be awardable multiple times.

...Or maybe this is too complex again...

It is a non-trivial problem rewarding constructive behaviour without also inducing unwanted and undesirable behaviour.

Answer 7

I like the idea of being able to either vote edits up/down or to provide some other means of quality control for edits themselves. I am relatively new to the Stack sites, and have already seen several posts "edited" with absolutely no change made at all (or trivial changes). It seems like users are really gaming the system just to get their "Copy Editor" or similar badges.