2

It's clear that the reviewers didn't understand this suggested edit.

It's a bug-fix. I had to add a comment to make it more than 6 characters though.

What can I do? I tried to edit it a second time as well.

  • Your 2nd edit is stackoverflow.com/review/suggested-edits/3231312 – psubsee2003 Oct 29 '13 at 14:34
  • 1
    Wondering: in your first suggestion you VERY nicely worked around that 6 character limit. Why did you just add "null6" in the second try? – Arjan Oct 29 '13 at 19:26
  • Relevant meta posts: #1, and #2 – jmac Oct 30 '13 at 6:10
  • @Arjan Because I didn't like the arbitrary restriction of 'minimum six chars'. My edit was 1 char, and that was perfect. By typing 'null6' I was trying to rebel against the system! Wooo! – Dan Bolser Oct 30 '13 at 15:50
15

Generally speaking, stay away from fixing bugs in answers with suggested edits. Most reviewers are not in a position to judge the context of your fix, or if the fix is correct.

Instead, comment on the post, and if you feel it's justified, downvote the post because the answer is incorrect.

You could even post your own answer to the question, with your version of the code, and demonstrate why your answer is correct. Then let the voting decide which answer is the better one.

  • 4
    Once you get to 2,000 rep, you can "fix" mistakes like these without the risk of a rejected edit. But, usually, by then, you do so extremely rarely. It is generally frowned upon unless you are a true expert and the error is very minor (and clearly unintended). – JDB Oct 29 '13 at 14:43
  • 4
    You shouldn't be making changes that violate the intent of the author, but when the intent of the author is clear and they have simply make a typo, fixing that typo is entirely appropriate. As for the correctness of the fix, a simple google search can easily validate the correctness of the comment in the revision history; no deep domain knowledge is needed here. – Servy Oct 29 '13 at 14:43
  • 2
    @Servy: good luck with convincing reviewers of that; the current reality is that reviewers will generally not know the problem domain or the context of the post and will not simply google to check if a change is correct. A small change in Fortran may be obvious to some, but I'd lack the domain knowledge to verify that, for example. – Martijn Pieters Oct 29 '13 at 14:50
  • @MartijnPieters That doesn't mean they're acting appropriately. The site policy makes it clear that this edit was appropriate, and that the reviewers improperly rejected it. Now, yes, I acknowledge that many edits of this type tend to be improperly rejected, just like I acknowledge that minor edits have a high probability of being improperly approved, but that doesn't mean that such improper reviews set site policy. See shog's answer on a similar situation. – Servy Oct 29 '13 at 14:53
  • To be fair, I saw this edit and skipped it... others could have as well? – ben is uǝq backwards Oct 29 '13 at 15:07
  • 1
    Fully agreed with @Servy on this, and invite people to look at this proposed-faq for code edits. Edits are clearly intended for fixing mistakes -- we don't get bent out of shape when someone suggests an edit to grammar, why would we correct the equivalent of a typo or misspelling in code in an answer? Learn to love the skip button – jmac Oct 30 '13 at 6:12
  • @jmac: Grammar and spelling mistakes are easier to verify, and the skip button doesn't get you badges.. – Martijn Pieters Oct 30 '13 at 8:47
  • @martijn, hoping that you are using sarcasm there... – jmac Oct 30 '13 at 9:25
  • @jmac: I am stating how I see this work in reality. Many reviewers will reject code changes out of hand, because they cannot verify if it is a typo fix or changes meaning. – Martijn Pieters Oct 30 '13 at 9:38
  • @Martijn, the way reality works is broken. It is wrong. And it results in lower quality. Rather than shrugging and saying, "it's how it works" we should be pushing to fix the problem with reviews instead. – jmac Oct 30 '13 at 23:45
  • @jmac: I didn't say I didn't want to see this fixed. I am merely stating that that is how it is right now. Someone asking for advise on how to have their suggested edits accepted cannot be told 'go fix reviewers rejecting code edits'. This reality may be depressing, may be wrong, but in this answer I am merely telling the OP that right now, with the status quo, it is better to not edit code, as that is likely to be rejected. – Martijn Pieters Oct 31 '13 at 12:45
3

Yes, the edit (at least the first one) was a proper edit and rejecting it was wrong (although not particularly unexpected). Most suggested edits to code are improper; seeing a good one is rather rare. This is proper, but because it's so rare many reviewers just don't look closely enough to realize that this is that rare exception, or are simply unaware that code edits can be appropriate under certain circumstances.

It's clearly just a typo. The text above the code snippet makes it clear that the intention of that post is to disable extglob, not enable it, and a quick search of the function makes it rather obvious that -s enables, rather than disables, an option.

The revision note even mentions this specifically, so the reviewers don't even have that as an excuse.

That said, when an edit of yours is rejected you shouldn't just suggest it again, that's not proper. Suggesting the same edit over and over until it's approved is abusing the system. When it was rejected you should have simply left a comment so that the author or another 2k+ rep user could edit the post.

You also need to make sure that if your edit is less than 6 characters that any other changes are good changes. The first edit was okay, the second was not.

I have since applied the proposed change, given that the authors intention was so clear.

  • Since I made the same comment to @ProgramFOX... I'll mention here... I don't necessarily agree that resubmitting a suggest edit if the first is rejected is necessarily bad, especially if it is a correct edit. Now doing it over and over again if it keeps getting rejected should make you stop and wonder "why". – psubsee2003 Oct 29 '13 at 14:46
  • @psubsee2003 So how many times do you think is appropriate, 3, 5, 10? Having a review be improperly rejected, from my experiences is quite rare. Virtually every time I've seen it has been someone proposing an improper edit over and over until it was approved by three people who didn't read it. – Servy Oct 29 '13 at 14:48
  • 1
    necessarily bad was a bad choice of words on my part. I probably should have had necessarily improper. I see the argument about not resubmitting, especially for people new to the site, but if you "know" the edit is correct (or have been told it was a good edit), I don't see the harm in trying a 2nd time before asking for editing help from a 2K user – psubsee2003 Oct 29 '13 at 14:52
  • 1
    @psubsee2003 The harm is that if you make a policy that it's appropriate to resubmit edits you'll see 99% of the resubmitted edits be edits that were rightfully rejected. And given the number of poor reviewers, a good number of those will end up being approved a second time, thus such a policy would cause more harm than it would help. – Servy Oct 29 '13 at 14:56
  • 1
    I strongly encourage having these improperly rejected edits being brought up on meta over and over until SE either prevents suggested edits on code entirely or they fix the robo-reviewing problem likely at the heart of this nonsense. – jmac Oct 30 '13 at 6:15
1

Executive Summary

Your edit was spot-on and should have been approved. This is an ongoing problem that keeps coming up because a large quantity of reviewers are allergic to effort and use botched heuristics to reject edits and lack of support for anything to solve the problem.

Your Edit was Correct

As stated in the Help Center:

Posts are Intended to be Collaboratively Edited

Editing is important for keeping questions and answers clear, relevant, and up-to-date. If you are not comfortable with the idea of your contributions being collaboratively edited by other trusted users, this may not be the site for you.

Edits should correct mistakes

Edits are expected to be substantial and to leave the post better than you found it. Common reasons for edits include:

  • To fix grammar and spelling mistakes
  • To clarify the meaning of the post (without changing that meaning)
  • To include additional information only found in comments, so all of the information relevant to the post is contained in one place
  • To correct minor mistakes or add updates as the post ages
  • To add related resources or hyperlinks

This is an on-going problem

Here are just three sample questions in recent history that have had the same issue:

What is even worse is that in your case you did everything right -- you even made a helpful edit comment explaining what the change was for!

People don't want to fix it

People will tell you that you should:

  1. Make a comment instead
  2. Make a separate answer instead
  3. Get 2000 reputation to fix the problem

None of these fix the problem. They create a separate problem. If I find a question on google that perfectly matches the issue I am facing, and the top answer doesn't work unless I fix it based on the comments below it, or find a separate answer with the same content and not that syntax error, then there is a problem.

I have seen absolutely no evidence that allowing suggested edits to code is more problematic than allowing suggested edits to grammar or spelling, and yet there seems to be a community consensus that there is some hidden problem here even from the mods.

At the same time, the simple solution of preventing <2k rep users from making edit to code gets heavily downvoted and people tend to agree with.

So we continue getting these sorts of requests from people who not only see what happens to their edits, care enough to question them, and then post a question on meta about it!

What can you do about it?

The only thing you can do right now is continue to make good edits, continue to get them rejected, and continue posting on meta about them, or to just wait until you have 2000 reputation and then make the exact same changes where curmudgeons reviewers won't see them.

The former has a chance of at least reminding folks that this is an issue that deserves attention, which the latter is probably the only way of actually correcting these glaring mistakes for future visitors. So I suggest doing both.

I also suggest a bit of sass toward people who suggest that correcting glaring typographical errors is a bad thing.

Note: Sass is probably ill-advised. Still feels good.

  • 1
    How long until we see a post with the exact title YARE (Yet Another Rejected Edit)? – John Dvorak Oct 30 '13 at 6:44
  • 1
    @Jan, we can take it upon ourselves to start editing the titles to any of these questions found as YARCE (Yet Another Rejected Code Edit). At least it will make them easy to find! Plus it rhymes with Farce, which is what the rejections are. – jmac Oct 30 '13 at 6:48
  • You can't have duplicate titles on SO, And I think it applies anywhere on SOFUE as well. – John Dvorak Oct 30 '13 at 6:51
  • @Jan, well then, this seems like a good case for using Roman Numerals! "YARCE (Yet Another Rejected Code Edit) I", "YARCE (Yet Another Rejected Code Edit) II", ..., "YARCE (Yet Another Rejected Code Edit) LIV", etc. And then we can start with "YARCE (Yet Another Rejected Code Edit) The Revenge", "YARCE (Yet Another Rejected Code Edit) The Revenge II", etc. – jmac Oct 30 '13 at 6:53
  • 1
    "YARCE (Yet Another Rejected Code Edit) - empire strikes back (hopefully)" – John Dvorak Oct 30 '13 at 7:00
  • @jmac Cool, give me a bounty then eh? I have other complaints about arbitrary score based restrictions... – Dan Bolser Oct 30 '13 at 15:52
  • @Dan, but then you wouldn't provide future examples of rejected suggested edits! And anyway, I am lacking in reputation on SO myself, if you want to come on over the The Workplace and make some good contributions however... – jmac Oct 30 '13 at 23:43
-4

It's an open secret that reviewing edits is a way to game the system. All the users with lots of points do this so they can crush new users.

  • 1
    How is this gaming the system? Also, this is somewhat funny because there is a major problem in the review system that edits are incorrectly approved; it's rather rare for edits to be incorrectly rejected. – Servy Oct 30 '13 at 15:58

You must log in to answer this question.

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