What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 127 Stack Exchange communities.

Original Statement:

I know Stack Overflow a bit (because of some of the other sites in the network). This is not really a question, but just for reference' sake: not everyone who makes minor edits is looking for reputation.

I made a little formatting edit here of which I know it is minor, but it is necessary because it messed up the code (the OP forgot a ' which makes the code run with an error!). This is not trivial. I had to add two little words for the edit to be large enough (and I did not want to ignore the mistake).

I think it is a bit lame to reject that edit (suggesting that "This edit is incorrect or an attempt to reply to or comment on the existing post."). Especially if the question is edited in the review queue after all. Either there is nothing to be edited or there is and in that case there is no reason to reject the edit. Why would you reject an edit and then do it yourself? Or am I misunderstanding something? Was the edit already made by the first edit reviewer? Hmm not sure here.

I am used to editing regularly on a site where I have +2000 rep, so it has become something automatic. Rejecting these edits does not encourage me to edit at all under 2000 rep.

Revised Statement:

From the answers below I retain that there are several principles for dealing with suggested edits:

  1. edits can be minor IF they deal with all the problems in a post. This explains why my edit was rejected, yet done manually by the reviewers. It does not explain why the people reviewing my edit did not also take care of the other problems in that answer.

  2. editors really have different views. Because I was criticized for dealing only with the code problem and not with the grammar issues in the question, I revisited the question, now also reviewing grammar. Again, two reviewers thought the edit was too minor. Three reviewers accepted it. (Interestingly, the ones that accepted it have a good accept/decline ratio, but I guess that is by accident :). In the comments on the present post, the grammar was called appalling, yet for others changing it is perceived as too minor. Bottom-line: reviewing edits is subjective and democratic.

  3. Many reviewers complain in the comments that minor edits needlessly take up their time and thus they tend to reject them. This is not communicated back directly to the editor unless he/she goes to look for it. The educative part of reviewing edit suggestions is underdeveloped. The editor should automatically be notified his edit was rejected and why. In my opinion, reviewers that edit a question based on a suggested edit, should also deal with all other problems in that post for precisely this reason. As such the editor can learn from the reviewer. After all, the overall quality of the site is in view.

I retract the following statement:

Certain reviewers have a 50/50 reject/approve ratio or worse. That is unfruitful (if not plain arrogant) and against the community policy [**edit**: there is no community policy on this, see Cody's answer].

It might be handy to check what the reject/approve ratio is for our most trusted users. A very quick check makes me believe that moderators, for instance, accept more than reject. But again, that is to be verified. As there is no hard and fast data, I retract that statement and apologize if it offended anyone.

For reviewers of suggested edits

Suggesting edits actually also takes time and effort. Please do not think that small edits that do not completely revise grammar and style (but that do fix a bug in the code) are somehow minor edits or attempts to gain reputation.

share|improve this question
14  
Everyone that is reviewing is donating their time. Don't waste that time with trivial edits. –  Martijn Pieters Aug 16 '13 at 8:41
3  
So let me get this straight: my 769/793 ratio on suggested edits is against community policy or arrogant? And here I though I was judging the edits on their merits. –  Bart Aug 16 '13 at 8:47
    
@Bart how did you get the raw? –  user221081 Aug 16 '13 at 8:49
1  
Find any suggested edit that a reviewer has voted on, click the "(more)" link, and you'll see their totals. Like this one of mine. Hey, look at that 73% approval! (Amazing, given all the bad edits I see popping up there in the queue.) –  Cody Gray Aug 16 '13 at 8:50
    
@Bart That's a poor ratio. You can draw the conclusions. I said it could indicate arrogance. Also, the whole things about editing costing time is besides the point. Everything on this website costs time. Deal with it. I DO agree with CodyGray's point that edits can be minor IF they deal with all the problems in a post. I have learnt something new... –  Private Aug 16 '13 at 8:54
2  
It's only a poor ratio if either the approvals or the rejections where somehow unjustified. –  Bart Aug 16 '13 at 8:55
    
51% for me i guess maybe 5% of all were vandalism, the other 45% too minor or invalid –  user221081 Aug 16 '13 at 8:59
    
Hm, my 462/612 ratio (43%) is probably because I typically skip posts that I'm not 100% sure about approving - I just want to help stop the robo-reviewers :P –  Doorknob Aug 16 '13 at 9:38
1  
@Private it can be if there are plenty of other things to fix as well. In your particular case I would have chosen to improve and accept. But I don't think it's too much to ask from an editor to check the entire post if you're going to suggest an edit anyway. Keep in mind that the whole review process also has an educational function. The more we can teach editors to do a complete job, the better it will be in the long run. Especially once they can make edits without review. –  Bart Aug 16 '13 at 9:47
1  
@Bart not many people actually go back and see if their edits were accepted or suggested I bet. –  Cole Johnson Aug 16 '13 at 9:48
1  
@ColeJohnson Do an inadequate job often enough and you can no longer suggest edits. That usually gets them to notice. Which is why I don't hesitate to be fairly strict in my reviewing. –  Bart Aug 16 '13 at 9:48
4  
@Private: It is too minor if you left any other problems with the post in place. If you are changing question code, you could be inadvertently masking the problem that needed solving. If you are changing the code in an answer, are you certain that you fixed it correctly? How can the reviewer know you fixed it correctly? –  Martijn Pieters Aug 16 '13 at 9:59
4  
@Private: The grammar of that post is rather appalling. Why did you leave that intact? The additional text in the comments looks like you tried to pad out your edit to meet the minimal length requirements, you should have addressed the other problems with the post instead. –  Martijn Pieters Aug 16 '13 at 10:03
2  
@Private: No, they are reviewing, and one of the options is to reject your edit as too minor. They'd be correct. You were wasting reviewer time here, why should the reviewer go out of their way to waste the time some more because you couldn't be bothered? –  Martijn Pieters Aug 16 '13 at 10:04
4  
"It does not explain why the people reviewing my edit did not also take care of the other problems in that answer* .... because you should have made a proper edit to begin with. That's not the reviewer's job. If they are so inclined to "improve" upon your edit, that is great. And it's a step I tend to take. But that is not the reviewer's job. –  Bart Aug 16 '13 at 10:11
show 12 more comments

3 Answers

Err, first of all, there is no clear "community policy" on this. You found one post that confirms your personal point of view and linked to it. The answer just below it, from the site's founder, with nearly the same score, and a 50 point bounty awarded to it, says something completely different.

Second, the reason we reject edits for being "too minor" has little to do with the reputation gain of the suggestor. Or at least, that's the furthest thing from my mind. Two of the more pressing concerns are:

  • The process of suggesting and reviewing edits is expensive, in terms of the number of trusted community members that must be involved to evaluate and vote on the edits. At a minimum, it requires 3 of us; sometimes more if the edits are contentious. That's 3+ people that could be doing something else to clean up the site. If you're not making a difference, you're effectively wasting our time, and we'll hit the "Reject" button as a deterrent.

  • The age of the question being edited, and whether it's worth bumping for a trivial edit that adds nothing of substance.

If you don't want to have your edits rejected as "too minor", there is a simple solution: fix everything you see that is wrong with the post. Rarely is a post so perfect that there is nothing else wrong with it but a single character out of place.. Dream bigger. Make your edit count.

That's part of the problem here, with the edit in question. I see several immediate problems with the post that you neglected to fix (despite hacking around the character limitations by adding some nonsense):

  1. The grammar is bad, and that makes it difficult to understand. Pronouns are missing, and so are commas.
  2. URL, CSV, and HTML are acronyms, and therefore always set in uppercase. They are lowercase here.
  3. Descriptive text preceding a code block should always be followed by punctuation (generally a colon or a comma).

Had you fixed all of those things, your edit would have easily been over the 6 character limit, and I would have definitely hit the "Approve" button.

In the form it was actually submitted, I would have used the "Improve" button to fix the stuff you didn't. But I probably still would have unchecked the "this edit was helpful" box, which would have caused the Community user to reject your edit.

Although I agree, the rejection reason that Rune FS picked was incorrect. DrYap's was a better choice.

I am used to editing regularly on a site where I have +2000 rep, so it has become something automatic. Rejecting these edits does not encourage me to edit at all under 2000 rep.

Well, the truth is, you shouldn't be making these excessively minor edits on other sites either. Although the "trivial edit" restrictions are relaxed because other users are not required to review your edits, the expectation still holds that if you're going to edit a post, you should fix all of the mistakes that you see.

share|improve this answer
2  
+1 for the age of the question being edited, and whether it's worth bumping especially. That is an angle too often overlooked. –  Martijn Pieters Aug 16 '13 at 8:43
    
I agree with most of what you state. I retain: fix all errors, not just one. You absolutely have a valid point there. –  Private Aug 16 '13 at 8:48
1  
I broadly agree with your principles, but this edit fixes the code to make it syntactically correct. It isn't too minor. Too minor is when an edit fixes one minor thing among many. –  Gilles Aug 16 '13 at 9:36
    
@Gilles Indeed, in fact, the code was incorrect without the edit. –  Private Aug 16 '13 at 9:39
1  
If the review system causes the reviewer to spend excessive amounts of time on minor edits, shouldn't the faulty system be fixed? Why don't reviewers get off their high horses and buck up a little? –  prusswan Aug 16 '13 at 9:41
2  
@prusswan There is no way around it. The system is not faulty. Reviewing takes time, at least if you do it right. The only alternative is to not review edits, but that will never work: there are far too many bad edits that would get through. –  Cody Gray Aug 16 '13 at 9:45
1  
@CodyGray If you take more than mere seconds to judge a spelling correction, you are taking too much time. while spelling corrections may not be your cup of tea, there are no ethical grounds for rejection. –  prusswan Aug 16 '13 at 9:53
1  
@prusswan The correction might take seconds to evaluate, but in a proper review you give the whole post a look as well, and see how that suggested edit works out in the post as a whole. That takes more than mere seconds. –  Bart Aug 16 '13 at 9:55
6  
@prusswan Great, I'll just start clicking the "Reject" button for all the edits you suggest. That will save me lots of time. But I'll continue paying close attention to other people's edits until they ask me to do otherwise. –  Cody Gray Aug 16 '13 at 9:59
    
@CodyGray and with that, you can kiss your future adminship chances goodbye. Take the black mark like a good boy –  prusswan Aug 16 '13 at 10:02
4  
@prusswan What does that mean? You are trying to tell me that I'm an idiot because I take time reviewing edits. I take offense to that. And I hardly think you can speak for everyone here, despite your attempts to do so. I think conscientious editors appreciate it when reviewers take their time assessing the quality of an edit, rather than just clicking one of the buttons on a gut call. I mean, I'm not just making that up, I have evidence: look at all the people who come to Meta to complain about one of their edits/flags being rejected. No one has done that yet for one I've reviewed. –  Cody Gray Aug 16 '13 at 10:07
    
I'm simply saying that bad reviewers need to stop dreaming that every minor edit is taking up 5 min of their time that they could use to save a dying person. I don't need to put up with your bullshit anymore but most users are simply not going to waste time fighting the rep they do not have. –  prusswan Aug 16 '13 at 10:14
4  
@prusswan No one said anything about saving a dying person. What I said was it was time that I could be spending cleaning up the site myself. And it isn't just me, it's at least two other people who could each be doing their own janitorial work. No one is forcing you to submit edits. If you don't like how the process works, you can feel free to abstain. I don't really know where all of this hostility is coming from, and why you're directing it at me. –  Cody Gray Aug 16 '13 at 10:21
2  
@CodyGray Another important point you didn't include is that while a suggested edit is pending the post can't be edited. Pending edits that are very minor, especially on recently posted content, can inhibit other users from posting more substantial edits while they wait for the post to be reviewed. Minor edits also consume the editors time, in addition to the reviewers, that could be better spent making valuable edits, and accepting them encourages users to suggest more of them; rejecting them encourages them to learn what we consider "valuable" edits to be. –  Servy Aug 16 '13 at 14:08
add comment

I would definitely have accepted this edit: it fixes a minor error in an answer. It's a typo in code. Sure, anyone who understands the answer would figure out the typo easily, but that's no reason to force every reader to go through this. This edit is not too minor because it fixes a major issue in the post. Generally speaking, edits that correct factual errors are not too minor.

The most common kind of too minor edit is an edit that fixes one or a handful of grammar, spelling or formatting issues while leaving many more. The boundary is obviously subjective to some extent; here's my take on it (for the common cases on Stack Overflow):

  • Correct a factual error → major.
  • Format blocks of code as code instead of text, or make unescaped HTML appear → major.
  • Make an identifier appear as code, remove code formatting use for emphasis, and other formatting issues that concern a single word or short phrase → minor.
  • One spelling or grammar correction → minor.
    May be requalified as major if correcting a word that is misspelled throughout and is important search fodder.
  • Add an important tag (e.g. the relevant programming language) → major.
  • Add or remove a secondary tag → minor.

When you suggest an edit, you should try to fix all the issues you spot. But if you don't have the time or competence for that, then do fix at least one major issue or a large proportion of the minor issues.

In this case, it would have been better if you'd fixed the grammar in the post, but nonetheless your edit was a good one and should have been accepted.


Note that while I agree with you on the right way to handle this particular suggested edit, your behavior in this meta thread is less than stellar.

Having a high acceptance rate for suggested edits is not the mark of a good reviewer. A “good accept/decline ratio” is not a high one. Accept ratios can vary depending on many factors including whether reviewers review organically or from the queue, what period of time they tend to review in, which suggestions they skip due to a lack of subject knowledge, etc. My own ratio is about 75% acceptance, but I've been doing this for a long time, and I believe that ratio includes my using Improve back when you couldn't mark a suggestion as unhelpful. I don't consider a 50% acceptance ratio any indication that the reviewer is doing a bad job. It does not point towards that reviewer being “unfruitful” and definitely not arrogant.

There are bad approvers and bad rejecters. The problem is symmetrical.

You argue that reviewers should fix all remaining issues in the post. This is backwards: if you're suggesting an edit, you're the one who's supposed to be fixing these issues. You're complaining that other people aren't doing your job for you!

Reviewing suggested edits does have an education value, and Decision on rejected edits should be displayed as a notification to the editor. But that doesn't change the decision that reviewers should take: if a suggestion is bad, it should be rejected, precisely for the education value. Indeed the education value is part of why suggested edits that are too minor are rejected, even though they do improve the post a little bit. (It isn't the only reason: other reasons include the facts that these edits bump the thread and that they move the post closer to community wiki status.)

share|improve this answer
    
For reference' sake I upvoted your answer. Can you explain, though, why you consider my behavior in this thread less than stellar? I am asking questions from the point of view of a relatively new user with low rep. –  Private Aug 16 '13 at 12:56
    
Yeah, I mean I basically agree with this in spirit, I'm just not sure about this particular edit. I guess it's an issue of where you draw the line on major vs. minor. Fixing the code certainly had some value, but there was so much else left untouched that I probably wouldn't have judged it sufficient. It is a bit of a dilemma though, at least for me, that a single misplaced ' is a substantive edit as long as it appears in a code block, but would not be if it were in e.g. the title of the question. What about fixing a misspelled word in the title, but leaving other problems—major or minor? –  Cody Gray Aug 17 '13 at 4:34
    
@CodyGray The consequence of this one missing character is code that doesn't work. That's a big consequence, so it isn't a minor edit. A single misspelling outside code is generally minor (see my “One spelling or grammar correction” bullet point). –  Gilles Aug 17 '13 at 10:39
add comment

While that may be true (and probably is for a lot of contributors), something to consider when making edits is that there are a lot of them happening, and that requires time from members to review them.

Those with a high enough rep can perform minor edits without needing approval, hence not taking other people's time with trivial (as in minor) edits.

share|improve this answer
    
That's just one extra reason to easily accept minor edits that add value to the question or answer. Cf. Should tiny edits be accepted or rejected in review. I mean, if you don't have time, don't spend it on reviewing... –  Private Aug 16 '13 at 8:40
1  
@Private "I mean, if you don't have time, don't spend it on reviewing.." NO. people don't want to spend time reviewing trivial edits. If you don't like having yr edits rejected, don't make them –  user223277 Aug 16 '13 at 11:52
1  
@Private to clarify, I make time to do and in some sites, to review edits, to 'give back' to SE for the assistance the site has given me. In that, I would rather be of better use for the community than dealing with small edits. –  user226423 Aug 16 '13 at 11:54
add comment

You must log in to answer this question.

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