7

The most recent example: https://stackoverflow.com/review/suggested-edits/14157557

This is not the first time somebody has edited one of my answers to fix a rather obvious bug, and then had that edit rejected.

My concern is twofold:

  • the rejectors don't know the language well enough to make corrections
  • the fixer is discouraged and stops trying to help

How do I bring such things to the moderators' attention so they can review the reviewers who made the bad rejects?

2
  • 3
    Code edits are really hard to judge. One one hand, the edit might fix a bug. On the other hand, it might introduce garbage. Generally I take the stance that code edits really shouldn't be done by other users. We don't want to put words (or code) into the OPs mouth.
    – JAL
    Commented Nov 1, 2016 at 19:32
  • Related (and possibly a duplicate): How can we be better at approving suggested edits that improve answers?
    – ale
    Commented Nov 2, 2016 at 19:40

1 Answer 1

8

Edits shouldn't be fixing code. Edits are for making posts more readable.

Bugs in code should be pointed out in comments so the author can fix them. This has been the guidance ever since editing was introduced. To reverse that would fly in the face of the cultural norms that have built up over years.

So, to answer your question, rejects of changes to code aren't "bad", so don't flag them. Just fix your code.

10
  • 1
    So do we have a reject type of "bugs in code should be pointed out to author in a comment"? Because the reason of "changes the original intent" is wrong -- my original intent was not to post buggy code. Commented Nov 1, 2016 at 19:35
  • There's the rub, though, innit? It's hard to judge whether you've got a minor bug in your code or it's a stylistic interpretation or taking advantage of an undocumented feature. But it's easy to judge if you've misspelled a word or munged a bulleted list.
    – ale
    Commented Nov 1, 2016 at 19:39
  • @EthanFurman it's close enough, I would use it as well. Users with edit privilege (2k rep on SO) can edit such code mistakes just fine, users without that privilege should only comment. Reviewers should not know about all the languages. Commented Nov 1, 2016 at 19:39
  • 1
    @AʟE.: Complete agreement on the misspelled word or bulleted list -- much less agreement on the others. Let me reiterate that it was an obvious bug -- at least if you know the language Commented Nov 1, 2016 at 20:27
  • 2
    @ShadowWizard: I must respectfully disagree, on both counts: I see a world of difference between "original intent" and "stupid bug"; also, I don't think folks should be rejecting posts on languages they are not competent in -- that's why there's a skip button. Commented Nov 1, 2016 at 20:30
  • 2
    @EthanFurman that's just how things work. No written rules, as this very answer explains that's the norm: suggested edits should not change code. Good or bad I can't say, but I don't plan to change it, and don't think it's worth the effort to try and change it for those who think it's wrong. Commented Nov 1, 2016 at 21:12
  • 1
    "Edits shouldn't be fixing code. Edits are for making posts more readable." No. Please edit code. Please. If a reviewer can't tell whether it's right or wrong they should skip.
    – hichris123
    Commented Nov 1, 2016 at 21:21
  • @hichris123 if everyone did that the edit would never be approved; if there is a bug in code, suggested by somebody, it can either be solved (if it's minor) or the answer should be down voted IMO
    – Ramhound
    Commented Nov 1, 2016 at 22:52
  • 2
    @Ramhound: minor bugs are exactly what I'm talking about - the suggested edit in this case was changing args[0] to doc in a four-line function. Anybody who couldn't tell that was a correct change should not be voting on it. Commented Nov 2, 2016 at 0:23
  • 1
    I'm simply conveying what the current norms vis-a-vis suggested edits of code are. If you want to argue that they should be changed, that's really something you should bring up at Meta Stack Overflow. Arguing about it in these comments will accomplish little.
    – ale
    Commented Nov 2, 2016 at 13:12

You must log in to answer this question.

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