6

I suggested an edit to a StackOverflow answer, which contained a bug in the code:

doc_rejoined = ''.join(doc) should have been doc_rejoined = ''.join(doclines)

I made the correction my edit and also improved the wording, which in my opinion was/is sloppy and informal. However, the community rejected the edit and approved a later edit where only the bug was corrected.

I'm asking because I'm a new user trying to contribute to the site. But if verbose edits aren't welcome I don't want to waste my time writing them. Should I restrict suggested edits to minor changes only? I want to contribute in a way that is desired by the community. What are the guidelines for editing posts?

  • 2
    Apart from the bit about amino acids (which is not relevant), I don't see an issue with the edit. – Oded Jan 14 '14 at 18:32
  • I think that's exactly the issue though @Oded. – Bart Jan 14 '14 at 18:38
  • I don't think any of the edit was appropriate, perhaps excepting the doc->doclines change (which does seem obvious). The original answer was sufficiently clear, and it's not like Martijn wouldn't have responded to comments. – Joe Jan 14 '14 at 19:00
  • @Oded I agree with @Joe; besides the amino acids thing (bad) and code correction (good), I don't think any of the changes made the answer worse, but I also don't think any of them made it better, and I wouldn't be comfortable approving an edit in which the majority of the changes were writing style tweaks where I didn't see any reason to prefer one version over the other. Change writing style if it brings clear improvements in correctness or clarity, sure, but I think post ownership should at least be respected enough that people don't make totally subjective style changes on a whim. – Mark Amery Jan 14 '14 at 22:18
  • 1
    For whatever reason, a large portion of the community feels the need to reject anything that touches code. They really shouldn't, but they do anyway. It most likely isn't about verbosity (save the amino acids bit), but rather about the edit to code. – jmac Jan 15 '14 at 2:08
  • Thanks all. Interestingly I only added the amino example because he had a statement to the affect of "but then all your lines would run together" and I wanted to present an example of that being desirable. But I was doubting the addition even when I made it; I should have gone with my gut. --These answers clear things up quite a bit, thank you. – binarysubstrate Jan 15 '14 at 15:56
5

But if verbose edits aren't welcome I don't want to waste my time writing them. Should I restrict suggested edits to minor changes only? I want to contribute in a way that is desired by the community.

Quite the contrary; suggested edits are supposed to be bigger changes that address issues in the post. We even have a reject reason for suggested edits that are "too minor."

The thing is, with edits, try not to add your own commentary in with the edit. My guess is that adding

(Which may be useful if processing amino acid sequences, and so on.)

is the main problem that the reviewers had with your suggested edit (I agree with Oded; the rest of the suggested edit looks fine.)


The other answer (since deleted) does bring up a point though; code fixes often will get rejected (if it's a substantial fix it will get rejected), though if it's a really blatantly obvious slip-up in someone's answer (like saying "pritn" instead of "print" somewhere), feel free to fix. Just throwing this as an addendum so that you know for later.

  • 3
    This. The code fix was clearly appropriate; the basically arbitrary writing style changes were not. I'll happily accept style changes of arguable value in the context of a large rewrite of prose which clearly improves the quality of the post overall, but none of the style changes here were clearly positive, and the amino acid sequences thing was just a weird and unnecessary example that I doubt many people would have any familiarity with. – Mark Amery Jan 14 '14 at 22:14
  • This helps. It seems like there is a fine line to walk on edits: editing minor code typos vs changing code, and editing spelling/grammar mistakes vs changing writing style. I'm a bit disappointed honestly that stylist changes aren't welcome (something you commonly see on Wikipedia for instance) because I think that would improve the site overall. But Wikipedia entries are anonymous as opposed to SO answers, so the difference/reasoning is easy enough to see. – binarysubstrate Jan 15 '14 at 16:03
-2

In general, answers shouldn't be edited for content - only for formatting issues. If someone else gives an answer, whether it's right or wrong, it should be their answer. If you think there is a mistake, you should either write a comment to the answer, or write your own answer. If you edit their answer, you are attributing to them something that is not what they said - and they might not agree with your change.

Even if it's a minor obvious typo, a comment is the way to go. Comments will get sent to the original answerer, and they can decide to incorporate it or not.

  • 4
    Fixing a minor obvious typo is entirely appropriate in an edit. Rejecting such a fix is an improper review. While commenting isn't wrong, (particularly if you are unsure if it's a mistake, or if you're not sure how to fix it) editing it when you can is better. – Servy Jan 14 '14 at 19:21
  • I don't think I agree with that. It's too easy to think you're fixing a typo and instead make a harmful change. While I would approve an edit that fixed a typo that was sufficiently obvious that a non-subject-matter expert wouldn't feel uncomfortable allowing it, it's easy to imagine a 'typo' that is an intentional difference. Particularly if it's a recent answer and/or an answer by a frequent poster (as in this case), it's easier to comment to make sure it's correct. – Joe Jan 14 '14 at 19:56
  • 2
    You may think otherwise, but the guidelines laid out for editing disagree with your opinion. – Servy Jan 14 '14 at 19:58
  • The guidelines are hardly specific; they don't differentiate between questions and answers, which the community clearly does from past discussion, for one thing. So much of 'typos' requires subject matter knowledge to understand, however; after all, if I don't know python, perhaps doc is a keyword that does something different from the variable doclines? – Joe Jan 14 '14 at 20:06
  • 4
    If you're unsure about an edit such as that then the appropriate review action is "skip" not "reject because I don't know". When you don't know whether an edit is appropriate or not, "skip" is the correct action. – Servy Jan 14 '14 at 20:08
  • Joe, this faq may help you understand the guidelines. Correcting small mistakes is clearly what the edit system is for. It is in the edit help page. It is on the about page. It is clear that comments are not for temporary information not for correcting mistakes. – jmac Jan 15 '14 at 2:10
  • Thanks Joe. Despite this answer generating some controversy I think it highlights the need to suggest changes/improvements in answer comments instead of making significant code or content changes (including writing style), which would be rejected. – binarysubstrate Jan 15 '14 at 16:07

You must log in to answer this question.

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