As I continue to assist with the moderation on Stack Overflow I have noticed that most people have a huge number of declined suggested edits.

I myself have probably only declined about 10% of all suggested edits I have reviewed, as I see most of them as improvements and in general only decline changes to the answer or question in cases where they change the intent of the poster.

Why are so users in general so though on suggested edits and should I in general take a more aggressive stance against edits?

  • 7
    Please don't abuse code formatting for elements that are not code. It's not a general highlighting tool. – Bart Mar 5 '13 at 17:24
  • 9
    A horrendous amount of suggested edits are either too minor or completely invalid - such as using code formatting for text that isn't code. >.> – J. Steen Mar 5 '13 at 17:24
  • Any suggested edit that abuses backticks like you just did should be rejected, btw. :-) – Martijn Pieters Mar 5 '13 at 17:24
  • 1
    Ah, sorry @Bart! I am still learning. – eandersson Mar 5 '13 at 17:25
  • 1
    @Bart Oscillating together... – J. Steen Mar 5 '13 at 17:25
  • And can you define 'most people'? – Martijn Pieters Mar 5 '13 at 17:25
  • @MartijnPieters Haha, yea exactly why I asked this. As clearly I am not doing something right. – eandersson Mar 5 '13 at 17:25
  • @Fuji No problem. Now you know. And a lot of my rejections are actually spent on users suggesting such edits. ;) – Bart Mar 5 '13 at 17:25
  • A lot of my rejections are spent on people just editing tags when there's so much else to fix in a post. That's too minor for me. – J. Steen Mar 5 '13 at 17:26
  • 8
    I also totally approve that you asked this. It means you're paying attention. We like people who pay attention. =) – J. Steen Mar 5 '13 at 17:27
  • 1
    I took a look at the first couple of pages of your review history and almost all the edits you approved had only approve votes from other users. In fact, many of the edits you rejected had some approve votes which would suggest the problem is exactly the opposite. – JJJ Mar 5 '13 at 17:37
  • Thanks @Juhana. Yea. I figured I was doing something wrong. I myself have been focusing mostly on approving edits that improve grammar, but I think I got a better idea now. – eandersson Mar 5 '13 at 17:40

I've rejected more than 50% of my edit reviews, probably because I give a larger than average value to authorship

  • I'll reject edits that add new content as invalid edit all day. You can post your own answer, or even leave a comment.
  • I'll reject edits that capitalize 2 letters and add a period as too minor all day. If it doesn't have a notable impact on readability, than you don't need to edit it.

on the other hand, I'll approve edits that

  • fix broken links
  • fix straight up errors in code
  • code formatting
  • spelling errors
  • grammar improvement if it actually makes the post more readable.

A good rule of thumb is

  • Is the post meaningfully better AND is it essentially the same post.

If that's true, I'd approve. If it's false, I'd reject or improve.

| improve this answer | |
  • I'll reject edits that add new content as invalid edit all day. I personally usually reject them, but sometimes, I'd approve if I find it appropriate. I do quite a number of of edits to straighten up the rest of the post, though. – nhahtdh Mar 5 '13 at 17:36
  • 1
    I'll reject edits that capitalize 2 letters and add a period as too minor all day. Capitalization and punctuation really effect readability. Although edits that leave a lot of other stuff to be fixed shouldn't be approved. – ɥʇǝS Mar 5 '13 at 18:18
  • @Seth punctuation can affect readability, if the punctuation error is in the right place. Capitalization, on the other hand doesn't affect readability very much at all. – Sam I am says Reinstate Monica Mar 5 '13 at 18:20
  • 1
    You're right it varies. I will add, lowercase 'i's hurt readability IMO. – ɥʇǝS Mar 5 '13 at 18:30
  • 1
    For edits that are too minor but are in the right direction, are there any other reasons for rejecting, or is it only to deny rep? – user000001 Mar 5 '13 at 18:33
  • @user000001 depends. does that edit actually make the post better? – Sam I am says Reinstate Monica Mar 5 '13 at 18:35
  • Like if the editor spots lets say 2-3 grammatical errors, and you cant see any that he missed. – user000001 Mar 5 '13 at 18:39
  • @user000001 do those grammatical error fixes actually make the post better? – Sam I am says Reinstate Monica Mar 5 '13 at 18:40
  • Probably not, since the post is comprehensible anyway. But they are correct. And this happens often I think. – user000001 Mar 5 '13 at 18:48

The answer from Sam is excellent. I'm a frequent complainer activist on this subject so I'll offer a perspective. In addition to what Sam said I reject edits that:

  • simply add backticks around keywords in the post but don't fix other obvious problems
  • add a tag but don't fix the whole question
  • do minor adjustments to code formatting that while not perfect was perfectly readable to begin with

Why the tough stance?

I want posts to be improved, with perfect spelling, grammar and code formatting. But with 2 rep points for each accepted edit too many people are submitting somewhat half-assed edits - while they may "fix" certain issues they don't fix everything - assuming that the "fixed" things needed "fixing".

My rejection rate is probably around 30-40%. I don't always reject, I also frequently improve - and I'm not slow to untick the Suggested edit was helpful checkbox if the author missed blatant problems.

I don't want to stop people making suggested edits, but I do want them to do a good job. Hopefully the attention to detail that they might learn in the process carries through to their coding.

| improve this answer | |
  • I didn't even consider that you get reputation for the edits. Thanks. – eandersson Mar 6 '13 at 8:04

The two other answers are excellent, but I wanted to distil out a detail in more memorable form, because I think that editors and reviewers often fail to realise this:

Your edits are not just judged by what you did, but also by what you didn't (do).

Correcting minor details whilst leaving major problems is frowned upon because it's antisocial:

  • It leaves the remaining work for someone else
  • It makes extra work for three reviewers
  • It's gaining rep 'on the cheap'
  • (It's being careless and lacking attention to detail, and most of us here are OCD :)

'Robo' corrections need to be avoided: a typical pattern is someone searches for a 'flag' word (like 'thanks') then goes through every single post just removing that word and not even looking at the rest. This is just rep farming.

Go on, tell me that unbalanced smiley isn't hurting your eyes...

| improve this answer | |

You must log in to answer this question.

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