5

I suggested this edit to an answer after evaluating both the problem and the solution with the right ingredients. I read question, answers, and comments carefully before I suggested the edit, because I wished to be sure not to introduce any consistency errors by editing. I think some of my reviewers didn't spend enough time as to really check this edit.

I don't try to suggest the same edit again. Instead, I commented the suggested edit below the answer.

Is this all I can do? (besides this meta question)

  • 4
    Your edit was good. I always appreciate it when others find a small error in my answers and correct it. It's always good to make answers better. – Lance Roberts Nov 25 '13 at 0:21
  • 4
    @LanceRoberts No, the edit was not good. It should have been a comment. – Doorknob Nov 25 '13 at 0:27
  • @Lance Edits should not change the content of an answer. Comments are to alert the OP of potential problems. – Aza Nov 25 '13 at 0:39
  • 5
    @Emracool, no, it's ok to make simple edits to fix simple errors. The goal is good answers, and it helps everyone, including the answerer who made the mistake. – Lance Roberts Nov 25 '13 at 0:46
  • 3
    If your answer is broken, then it is a poor answer that does not solve the problem. If someone like Wolf wants to fix it, that improves the quality of the answer and should be encouraged. Good users who are able to proofread code are being given the wrong message every single day because reviewers are allergic to the skip button based on some sort deluded concept of ownership. If you don't want your answers edited, don't make mistakes. If you don't like an edit, roll it back. Rejecting it out of pride is anti-collaboration and goes against the SE concept. – jmac Nov 25 '13 at 0:48
  • 1
    @Lance Well, that argument really doesn't hold water. It helps the answerer more to either change it themself. A comment makes them aware of the answer's potential error. Also, editing answers risks inducing further errors which are not the fault of the answerer. – Aza Nov 25 '13 at 0:48
  • 2
    @jmac There are two problems with this: First, it's not your answer to edit. You could be unintentionally introducing errors which are not the fault of the answerer. Second, the same effect is achieved by commenting, but without the risk, and with the responsibility of the person whose rep is at stake to begin with. – Aza Nov 25 '13 at 0:49
  • @Emracool, there are many times when the answerer doesn't pay attention to the comments and doesn't come back. Meanwhile many others might be trying to use that answer. All edits are risky. – Lance Roberts Nov 25 '13 at 0:50
  • 4
    @Emra, perhaps you haven't read our help center on editing: "if you see questions or answers that can be improved, you can edit them. Use edits to fix mistakes, improve formatting, or clarify the meaning of a post." It is my answer to edit. Along with anyone else in the community who feels the need. The effect is not achieved by commenting since comments are second-class citizens and if accurate should be edited in to the post. – jmac Nov 25 '13 at 0:51
  • @jmac "and if accurate". The community managers have stated (though I can't remember where) that the intent behind "use edits to fix mistakes" is to correct things like single-character typos. Something which significantly changes the function of the answer in a way which may not be correct is not a good edit, even if it may make the answer correct. – Aza Nov 25 '13 at 0:53
  • 2
    @Emra, if you don't know if it's correct learn to love that skip button as Shog9 has said. And <2000 rep users cannot make one character edits. At least get your ducks in a line before you start firing. – jmac Nov 25 '13 at 0:55
  • @jmac I use the skip button. Most other reviewers do not. Plus, people push single character edits through by changing words around in minor and unnecessary ways. – Aza Nov 25 '13 at 0:56
  • My point is that the reviewers probably don't know whether the edit is correct, and aren't in a position to judge it. Yes, ideally they should use the skip button, but pragmatically they don't, so we have to accommodate their existence. – Aza Nov 25 '13 at 0:58
12

This edit was validly rejected because it changed the directory in the answer. Remember, the answer is the solution provided by a user, and is representative of their work. You can't go and change that, because the answer would no longer represent their solution.

In this case, the reason:

This edit is incorrect or an attempt to reply to or comment on the existing post.

is somewhat unclear. If you've spotted what you believe is an error in the answer, then a comment is the correct course of action.


To clarify: Reviewers should not have to gauge the accuracy of an edit. In doing so, this would violate the purpose of reviewing, which is to create objectively good answers. While yes, in an ideal world, people would skip over edits whose accuracy they could not gauge, this does not happen (at least on Stack Overflow), and so we must create pragmatic workarounds by disallowing questionable content edits.

The key word here is "questionable": if your edit is obviously correct, such as a small typographical error in an answer (like bra instead of bar), it is okay to edit. However, if your edit could potentially change the accuracy of an answer, then it will be rejected.

  • 1
    I think it's worth adding that if your comment is ignored, you can always post a better answer yourself. – sashkello Nov 25 '13 at 1:10
  • 3
    @sash, that is a horrible alternative. So we get two answers, identical content, only one works, and one doesn't, because of a missing bracket? How is the community supposed to evaluate which is right? Momentum will have the first answer which is 99% right get more votes, while the alternative below the fold gets ignored because nobody can tell the difference. Why are you proposing making the correct answer more difficult to find? – jmac Nov 25 '13 at 1:20
  • @jmac 1. Missing bracket is "correct minor mistakes" and should be an edit. 2. Don't jump between things - either talk about this particular question (major edit), or in generalities. Have you actually read what you wrote? It sounds ridiculous "one works, and one doesn't" "nobody can tell the difference" - isn't it the purpose of SO to post answers and for users to be able to evaluate it themselves? There is a voting system for that. Otherwise let's post one answer and edit it all the time until it is perfect. – sashkello Nov 25 '13 at 1:24
  • 2
    @sash, the purpose of SE, as was stated at its creation, is to make answers easier to find. That is why edits are allowed on other people's content. If I dig up an answer from a year ago that doesn't work as written because they forgot /bin/Debug at the end of the example, then are you really advocating I go and post a second answer rather than edit the +20 accepted answer that already exists? That will not achieve the desired results (my answer would be buried, and would likely be flagged and deleted for covering what was already discussed in another answer, which is also discouraged) – jmac Nov 25 '13 at 1:27
  • 1
    @jmac That's what a comment is for. Serves both as a flag to the OP and future visitors. – Aza Nov 25 '13 at 1:34
  • @jmac Don't rant about the purpose of SE. This is meaningless. Read the rules - if you disagree, propose to change them. But don't encourage people to misuse the system because you think it is better. – sashkello Nov 25 '13 at 1:40
  • @jmac About this particular example. Comment and leave it there. It will not go anywhere - people will read it and get the idea. If the answer is wrong, its score must reflect it. If not, then why changing it? People voted for the original answer, not an edited version which might not work any more. – sashkello Nov 25 '13 at 1:41
  • This is true, because of the order of reasons I used in the comment. If did better omitting the first. What I learned from this is that the comment-to-edit relation is crucial. Thanks for pointing me on this. – Wolf Nov 25 '13 at 5:58
  • @Josh Actually, the translation of the question read like an actual question. I don't think it was indicative of spam. Plus, since the user actually deleted their own question, I think they aren't a spambot. – Aza Nov 25 '13 at 7:38
  • 2
    Your answer and comments are incorrect on multiple counts. Please read the purpose of comments: “Comments are temporary "Post-It" notes”. Correcting a mistake in an answer is not a temporary thing, it does not belong in a comment. Reviewers should definitely gauge the accuracy of edits, it's part of the job. As a reviewer, if you don't know, skip. @sashkello This goes for you too: you are pretending to instruct people on SE's rules when you clearly haven't read them, please do read them. – Gilles 'SO- stop being evil' Nov 25 '13 at 10:27
  • 1
    @Gilles Please read the discussions below and above. Comments are used for clarification and to correct mistakes. Reviewers should not gauge the accuracy of an edit - this is policy on SO. The "if you don't know, skip" rule is great in theory, but doesn't work on a site like SO where people review because there are reviews to do. The edit rule for "fixing minor errors" is intended to apply to small things like typos, and not anything that changes the functionality of a post. – Aza Nov 25 '13 at 16:39
  • @Emracool I have read the discussion. I have also read the rules, and I have a lot of experience reviewing, including back when there was a sore lack of reviewers on SO (which is no longer the case). If you don't know, skip. The fact that some people are doing it wrong is no reason for you to also do it wrong. Comments are not “used for clarification and to correct mistakes”, they are used to “request clarification” and to “Add relevant but minor or transient information to a post”. – Gilles 'SO- stop being evil' Nov 25 '13 at 17:29
7

Background

This issue comes up very frequently:

Suggested Edit rejected because reviewers didn't know programming language
Why was my code edit rejected?
Why was my suggested edit rejected repeatedly?

What Do the Rules Say?

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.

Useful Information in Comments Should be Edited in to the Post

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

Comments are not Permanent

What are comments?

Comments are temporary "Post-It" notes left on a question or answer. They can be up-voted (but not down-voted) and flagged, but do not generate reputation. There's no revision history, and when they are deleted they're gone for good.

Comments are Not Intended to Correct Mistakes

When shouldn't I comment?

Comments are not recommended for any of the following:

  • Suggesting corrections that don't fundamentally change the meaning of the post; instead, make or suggest an edit;

Conclusion

Code edits are often rejected because people are too lazy to properly evaluate whether or not they are correct, and prefer clicking "reject" rather than "skip" if they don't have the inclination to properly review. To get these sorts of edits approved, you need to bend over backwards to make your average reviewer's life as easy as possible:

Make sure the edit to code is appropriate

I put together a handy to cover when to edit/not to edit code:

https://meta.stackexchange.com/questions/200336/when-should-i-make-edits-to-code

Improve your suggested edit comment

Right now you wrote:

fixed some minor issues, the default settings produce a Debug version

You can improve this to:

This will not work because the directory is wrong, so /HelloWorld/ was changed to /HelloWorld/bin/Debug

Or something more descriptive that tells the reviewer, "This is clearly wrong, and this is why" with a glance. Otherwise they will just reject it because it's "safer".

Wait for 2000 Rep

Because reviewers are so horrible at these sorts of suggested edits, you can just wait until 2000 reputation so you don't have to deal with them when making common-sense improvements.


Have faith knowing that you're in the right. Keep fighting the good fight.

  • 3
    Your answer is spot on. – Lance Roberts Nov 25 '13 at 0:47
  • 1
    I really don't think the solution to this problem is "ignore quorum, since we're right anyway." Carrying the site in multiple directions is not beneficial for anyone, and creates further problems in communication and action. – Aza Nov 25 '13 at 0:50
  • 4
    @Emra, on my side: Help Center, SO concept, Answer Quality. On your side: Confused Sense of Ownership, Complex workflows to get worse results, encouraging reviewer laziness. I agree that taking the site in two ways is stupid, I just refuse to sacrifice quality for laziness. – jmac Nov 25 '13 at 0:53
  • 2
    @jmac Help Center disagrees with you; "fixing errors" is intended for minor things like single-character typos, not things which could make the answer incorrect. I don't know what you mean by "SO concept." Answer quality is improved by commenting because the answerer can edit the answer to reflect its accuracy. I'm not sacrificing quality for laziness. This is to ensure quality of answers. – Aza Nov 25 '13 at 0:55
  • 2
    @Emra, you seem to misunderstand the SE concept. The content belongs to the community -- not to you. If I find a mistake in your grammar, I fix it to make the answer better. You are not the guardian of your own content, and you are not allowed to refuse to allow edits because you are wrong and stubborn. That hurts answer quality. The community is smarter than you are. That is why Stack Exchange works. Refusing to let the community touch your answers makes no sense. – jmac Nov 25 '13 at 0:58
  • 1
    @jmac I'm pretty sure accuracy trumps ownership. When did this become about me? I've never rejected an edit from a comment I thought was correct, and I almost always skip past code edits I don't understand. Please stop targeting this at me. We should talk about the general case; if you insist on making this about me, I will leave. – Aza Nov 25 '13 at 1:00
  • 1
    Totally disagree with this. "to correct minor mistakes or add addendums / updates as the post ages" doesn't mean you need to make a correct answer out of incorrect answer (or in some cases vice versa). Edit reviews shouldn't evaluate the correctness of answers, this is not their purpose. If you disagree propose a change to the SO edit rules rather than calling people who exactly follow the protocol lazy. – sashkello Nov 25 '13 at 1:06
  • 4
    Fixing other people's code = bad. You don't know what the OP intended, and they are not informed about the edit in an obvious way for quality control. Code corrections are what comments are there for. – Pekka supports GoFundMonica Nov 25 '13 at 1:15
  • 2
    This has been gone over already in the linked post. To make an absurdly brainless example, if someone means to type SELECT foo JOIN bar but accidentally writes SLEECT foo JONI bar then fixing it is not 'evaluating the correctness of answers' but along the lines of correcting a typo in the intended answer. If the answer doesn't work as written, why would we leave it as is? That strikes me as just about the most counter-productive stance to take when trying to help people get proper answers. – jmac Nov 25 '13 at 1:17
  • 2
    @sashkello, so why are you saying that /HelloWorld/ to /HelloWorld/bin/Debug is a different answer entirely? Have you tested both? Does the first work just fine, and the second indicate something entirely different? You are seemingly suggesting that this is some radical change to the code different from SELECT vs. SLEECT, but if a reasonable user versed in that field (as Wolf would seem to be) thinks it is a good change, what is your justification for invalidating his expertise exactly? – jmac Nov 25 '13 at 1:33
  • 2
    @sash, I absolutely refuse to concede that any code posted as an answer is intended to not work properly as written without something written to explain that it is pseudo-code or not tested. And if someone turns that pseudo-code or non-working code in to something that works as written, no matter how major it looks it should be a part of the answer. Our goal is provide reliable answers that are easy to find, not to prevent people from improving things because of how they look. – jmac Nov 25 '13 at 1:44
  • 2
    @jmac You are absolutely refusing to concede a point nobody is trying to make. You are misconstruing the arguments you're responding to a la straw man, and responding as if they said something similar to what they said but which they did not say. – doppelgreener Nov 25 '13 at 1:46
  • 2
    @sash, I have written a large explanation open to community editing explaining exactly what I believe that rules on editing code for users under <2k should be. I have also written an answer explaining exactly why the alternatives proposed here do not correspond with the site rules. My stance on this is crystal clear -- productive edits that make better answers should be encouraged because that is what the rules say edits are for. – jmac Nov 25 '13 at 1:58
  • 2
    @jmac Your large explanation says under "Answers": "Don't ... Change the code logic or functionality - even if you think you're correcting it" (this Don't was there in your original version, though I added the "even if you think you're correcting it" myself afterwards) – doppelgreener Nov 25 '13 at 1:59
  • 2
    Two parting comments. The rules are principles. Our goal is to get high quality answers. Editing helps. That's why it's encouraged on other people's answers (see the about page). If the rules truly prevent you from improving the encyclopedia, ignore them. At the end of the day, broken code helps nobody, and if people want to fix broken answers they should be encouraged, not downvoted on meta for bringing it up. – jmac Nov 25 '13 at 2:23

You must log in to answer this question.

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