Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 153 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

I'm new to reviewing edits on SO and I'm conflicted about approving or rejecting formatting-only (or mostly) edits like this one or this one. Personally, I'd like to approve any edit that improves the post, even if it's just a couple of words.

I've read Are edits that only change formatting in text appropriate? and Why isn't a format only edit allowed? and SO doesn't accept edits that only address formatting and they are leading me to be more permissive in accepting edits when the edits cover all the obvious problems with the post, but I don't feel they give a full (and consistent) explanation of the rationale behind this. So my two-fold question:

  1. What is the benefit and purpose of rejecting edits as too minor?
  2. What is the harm in accepting formatting-only edits?
share|improve this question
If it's all there is to fix, there is no problem in accepting something minor which improves the question. If it glances over several other issues however, it might be too minor. Rejecting them would instruct the editor to make better and complete edits. – Bart May 12 '13 at 23:49
Reviews take time. We don't want to use the precious time of users reviewing minor edits. Accepting minor edits makes the posts better - no doubt. But I does not improve the editor. – juergen d May 13 '13 at 0:04
FWIW, while apparently part of a small minority, I pretty strongly disagree with the whole idea of an edit as "too minor". I don't find the arguments in favor of "too minor" at all compelling. But as this is a community decision, my opinion is "too minor" to matter :). – Ben Lee May 13 '13 at 0:19

So...since I was on both of those reviews (small world, huh?), let me walk you through a bit of my thought process.

The first post - there wasn't much that needed improving. There weren't any misspellings, grammatical improvements that could be done, or anything that was glaringly added to the original question - but the formatting could be changed. That, I'm fine with.

The second post - which I'm surprised got approved - I rejected as too minor, because it didn't address the other obvious misspellings or grammatical errors in the post.

Now, to your point:

What is the benefit and purpose of rejecting edits as too minor?

An edit should improve the quality of the post, such that it is clear when someone else comes in to gain knowledge from it. Edits that only fix one or two things while overlooking other, obvious problems are, to me, too minor.

We want to improve the quality of questions and answers, so saying "no" to the minor revisions is one way to do that.

What is the harm in accepting formatting-only edits?

There's no harm in it. But personally, I want to see more. If there's really nothing left to improve, then it's fine (as in the first example), but if you can improve more, please do so.

share|improve this answer
I agree, just want to add another parameter I take into account: post age. Keeping in mind that edit bumps the question to the front page, it's not something to be ignored. – Shadow Wizard May 13 '13 at 6:13

I agree with most of the points in Makoto's answer. However, regarding one claim:

There's no harm in it

In my opinion, incomplete edits are actively harmful, and not "helpful, but just barely".

  1. Approving incomplete edits incentivises a quick, cursory reading of a post, followed by a superficial edit. Since there is a rep reward for successful suggested edits, a series of crappy incomplete edits becomes more profitable than thorough editing.
  2. It is wastes reviewer time, since someone needs to click improve and basically make a proper edit from scratch.
  3. Perhaps most egregiously, it often interrupts another user's more thorough edit, becoming a nuisance to someone who is doing their due diligence in improving the post

None of these arguments apply to posts where there are no other problems besides a formatting error, of course, but approving incomplete edits is actually harmful.

share|improve this answer
Whoa, I don't remember cosigning on incomplete edits. I absolutely agree that approving incomplete edits is bad, especially because it enforces the behavior, and that the post is no better off than if it weren't edited. – Makoto May 13 '13 at 0:50
@Makoto I'm not saying you approve of incomplete edits. My point is this: it isn't just that people who make incomplete edits aren't doing enough; their net contribution is actually negative. – Asad Saeeduddin May 13 '13 at 0:52
We're on the same wavelength then. – Makoto May 13 '13 at 0:53
In fairness, argument 3 doesn't really count since the system allows a 'more substantial edit' to be applied on top of a smaller edit, based on amount changed. Fully agree with 1 and 2 though, and 3 half for this reason. – Niels Keurentjes May 13 '13 at 0:56
@Niels The metric for "substantialness" is extremely fickle. It relies on the number of characters changed, and you have to add at least 25 characters more than the clashing edit to get through. – Asad Saeeduddin May 13 '13 at 1:04

To answer your specific examples:

  • I would have rejected this edit, it makes some improvement but totally fails to fix the other capitalisation and grammatical errors. This means the question may still be a candidate for the low quality review queue. When a question is only 3 sentences long there is no excuse for not doing a proper fix on it. Those with sufficient rep could also opt to Improve the edit, and deselect the This edit was helpful checkbox so the original editor doesn't get the rep for an accepted edit.
  • this edit is harder. Did the edit improve the question? It did but only a little. Did it reduce the quality of the question? Not really. I would probably accept this edit, although reluctantly - while the suggested edit does have to be actioned, the editor could have found far worse questions that need attention.

There is no harm in accepting formatting only edits, but we don't want to make it a free-for-all gravy-train, where even the tiniest "formatting adjustment" gets accepted - that isn't the purpose of the feature.

share|improve this answer
"the editor could have found far worse questions that need attention" you could say that about nearly every edit! If it makes an ok answer/questions good (or good one great) then format only can be a good thing. – hayd May 13 '13 at 10:44

I personally would've rejected both as 'too minor'. Editing someone's post is invasive, sometimes even considered offensive, so better make it worth it so the original poster also sees the added value in his work being adapted. Given the amount of questions from non-native speakers, it could even be considered degrading to just fix a few minor grammar mistakes after they spent an hour making it readable and understandable.

My rule of thumb for edits is that they need to substantially improve the post. Just indenting 3 lines of code because they should be according to someone's personal coding standards is not an improvement, that's nitpicking. Indenting the same 3 lines of code because it suddenly makes the code understandable or logical most certainly is a substantial improvement.

The second example you mention is actually doubly troublesome since the editor decided to clog the review system and five people with nitpicking over 2 words that scream 'non-native speaker' but didn't make the post unreadable or bad to any extent, and he and 3 approving editors didn't bother to fix the other bunch of grammar mistakes. Fix all or fix none I'd say - the end result was nothing better than the original post this way so the edit shouldn't have been made nor approved.

share|improve this answer
Okay. Invasive I can see...but offensive? That's a stretch for me. By and large, you're right in that there are a lot of non-native English speakers out there, and they do try their best to make their question more readable. That's great! But if there are things to be improved, I don't see the harm or offense in improving it. That's why we have this system. – Makoto May 13 '13 at 0:29
@Makoto It's quite uncommon, but I've seen people take offense at people editing their posts. It usually starts with a "You changed my precious words? How rude!" comment and goes downhill from there... <sigh> – Yannis May 13 '13 at 0:41
@Makoto if someone spent an hour trying to do something right, and then finally pressed 'Submit' when they thought it was, only to have someone change an apostrophe and 2 spaces within 2 minutes, I can't always blame them for being offended. That's why I think your edit should also convince the original poster that it was an actual improvement, not just nitpicking about details such as the common code inlining of a function name in text which I also always reject if it's the only change. – Niels Keurentjes May 13 '13 at 0:45

You must log in to answer this question.

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