8

I found an answer that was helpful but had a syntax error in the path. I made the correction (so that the answer as-written would at least be accurate in this regard) and it got rejected apparently out of hand. I resubmitted the same correction (thinking perhaps that was in error) and again it was rejected:

https://superuser.com/review/suggested-edits/483842

This edit deviates from the original intent of the post. Even edits that must make drastic changes should strive to preserve the goals of the post's owner.

  • How does correcting an error in a system path deviate from the original intent of the post? It clearly wasn't intended to be incorrect or an example of a bogus path.

This edit was intended to address the author of the post and makes no sense as an edit. It should have been written as a comment or an answer.

  • What? The author crafted a fine answer, it just had a typo. This correction was made to fix a syntax error, not editorialize on it. (This was given as a reason in the first rejection as well).

I'm sure there are many spurious or bogus edits here that need to be filtered out on a daily basis, but it's as if nobody bothered to even read the corrections I submitted. This isn't my first time making edits on Stack Exchange - I like to contribute even in this small but useful way but it's demotivating when the most straightforward of corrections is summarily rejected.

As a side-note, I also corrected a hyphen ("per-user") because my original path correction didn't contain six or more characters. That was way down at the bottom of the priority list of Things to Bother Correcting but still, I bothered just to ensure the syntax error correction got submitted.

7

An extract from this answer What is the correct way to deal with obsolete answers? by slhck (a SU moderator) provides some clues (my emphasis):

The ideal solution in such a case, of course, is to edit and correct the answer if you can.

Since our editing guidelines say that you should keep edits substantial, but not radically change the meaning of a post, we welcome any edits that correct obvious mistakes, but don't rewrite the entire answer if possible.

If you suggest an edit for improvement and it gets rejected, try and see if the OP is still active so you can ask them to fix their post with your help.

You could also come to Meta when you spot a bunch of outdated and wrong answers.

If editing is not possible or would require a major rewrite of the post, you should rather downvote and comment. Explain what exactly is wrong with the post and how to possibly fix it.

The above quote was from an answer referring to obsolete answers, but I believe it can also apply to answers containing an error.

In this particular case, there is supporting evidence for your change:

So I've gone ahead and made the edit for you.

  • Normally you should add a comment for the OP, pointing out the error and allow him to make the change himself.

  • If, after some time has passed, the OP doesn't respond then go ahead and propose the change.

  • When you have sufficient reputation (2000) you can make the change directly yourself without requiring approval.

  • Thank you - in the future I'll keep that in mind (I jumped to edit given the age of the answer). – MartyMacGyver Jan 8 '16 at 18:52
  • @MartyMacGyver Note that is this case, although it was an old question, the author was "Last seen Dec 5 '15 at 19:18" (from his profile page). – DavidPostill Jan 8 '16 at 18:57
  • 1
    Indeed. As noted in my comments on nKn's answer here, I went to edit because this seemed to fall within the umbrella of quick correction versus needing a longer discussion with the author. Unless they're online within the last few days chances are they won't be replying for a while, and I'm not sure how long one wants to wait to fix a technical typo once discovered. It was an obvious mistake, and while it was an old answer it wasn't obsolete - I don't think the path given was ever valid (so far as I've found). – MartyMacGyver Jan 8 '16 at 19:02
2

I'll move this from a comment to a separate answer because it's really a different point than the other two answers.

Edits can be used to correct unintentional errors, like typos. but generally not to change technical details of an answer, even if they're wrong. If you think an answer is fundamentally incorrect, add a comment, downvote, and/or write your own answer.

Distinguishing between typo and the author being misinformed is where many edits fall short. Reviewers often don't know all of the technical details, and shouldn't be in the role of determining who's right. When in doubt, reviewers err on the side of the OP. You need to demonstrate that you're fixing a typo rather than changing the OP's technical details.

The first question the reviewer asks is whether the edit could be fixing a typo. In your example, leaving out the /chron parent directory is the kind of thing an author could do unintentionally. So the next question is whether the proposed edit is correct; there's the rub.

Solution: Include a link to an authoritative source in the edit comment, or add a comment on the post with citations, then refer to that comment in the edit comment. Notice that DavidPostill did that in his answer here.

0

In my opinion, these rejects are correct. Even if you are 10000% sure that the path is wrong, you shouldn't use the Edit button for that.

Answers and questions themselves are each author's "property", and edits should be used to improve readability, correct typos, tags, formatting but not any relevant content/technical part.

In the case you describe, you have several options:

  • Write a comment to the user's post and tell them the correct path and encourage them to modify their post to fix it. Feel free to downvote the answer if you think it deserves it.

  • You can write an answer with the correct content. You can even mention that you're extending @foo's answer to improve what you think is inaccurate.

  • 1
    Fair enough - I see your point. I am more likely to comment on an answer when the dialog is current (versus submitting an edit to a several-years-old answer) but I'll consider that next time - your point here makes much more sense than "makes no sense as an edit" or "deviates" and had that been one of the reasons I might never have written this meta post. – MartyMacGyver Jan 8 '16 at 18:48
  • 1
    That said, I did correct a typo here, which is in the domain of editing as you describe (it's a typo in a path, which is itself a technical part, but this was clearly not intended include such a typo and an edit seemed an expedient way to make the correction). I certainly wouldn't downvote them because of that, and submitting an identical answer just to correct an old typo simply adds to the noise. – MartyMacGyver Jan 8 '16 at 18:57
  • Indeed. I barely use downvotes in these cases (although one doesn't have to provide a reason for it), even if the OP doesn't want to change their answer. I usually leave a comment pointing the mistake and it's up to them if they want to change it. – nKn Jan 8 '16 at 19:01
  • 1
    I keep looking at the "property" aspect of your answer - which comes first here: accuracy or proprietorship? My experience on Stack Exchange has been one of accuracy first where matters of fact are concerned (e.g., correcting a typo - technical or not, versus correcting an opinion or a counter-example meant to have a typo). This doesn't come up a lot (usually there's something more worthy of debate with the author than a small, old typo) but I wouldn't consider it untoward if someone edited a typo in my answer - in fact, I'd be surprised if they didn't. – MartyMacGyver Jan 8 '16 at 19:10
  • Don't take it too literally either, I meant that each one is responsible for their posts and should be the ones that decide what content they wish to publish. I guess accuracy is fairly measured by the upvotes (and user's reputation after all), usually the more upvotes mean the more accuracy, quality and usefulness. Edits may help to this three attributes, but should not interfere with the content itself. – nKn Jan 8 '16 at 19:22
  • Understood - I agree. – MartyMacGyver Jan 8 '16 at 21:10
  • In general, I add a comment to the post and state my case. If this is a newish question or answer, then the person who posted it will mostly likely see it and take action. Heck, the question doesn’t need to be old as much as you just need to be sure someone is paying attention. If you are still concerned and have no clear indication someone will pay attention, post a new answer that references the original answer and corrects the issue. But I would never edit the content directly like that without the original poster’s approval. – JakeGould Jan 10 '16 at 4:17
  • It'd be better to get the content of an upvoted / accepted answer fixed in the first instance. If the author comes and edits back later, that's fine. Random users trying to apply the answer to fix their problem are more important than the author's sense of ownership. – Sir Adelaide Feb 13 '17 at 3:40

You must log in to answer this question.

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