What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 131 Stack Exchange communities.

I was trying to correct minor mistakes in this post.

Which I understand is allowed in the Help Section.

Some common reasons to edit are:

  • to fix grammatical or spelling mistakes
  • to clarify the meaning of a post without changing it
  • to correct minor mistakes or add addendums / updates as the post ages
  • to add related resources or hyperlinks

The edit has been rejected, and since it has been rejected a second time with the reason "Ask on meta, don't just try to override the mod", which I completely agree with, here am I on meta.

Could somebody give me an hint?

share|improve this question
3  
Minor mistakes unfortunately (well, reasonably actually) doesn't include code. Don't edit that, leave a comment. –  Bart Jul 18 '13 at 19:37
    
Ok, I didnt' understand that. –  Jean-Claude Arbaut Jul 18 '13 at 19:38
6  
You're not the only one. It should really be clearly stated in the Help Center. –  Bart Jul 18 '13 at 19:38
2  
Keep in mind that "minor" isn't just about the number of characters edited. Even the smallest changes in code, from the point of view of character count, can have an enormous impact on that code, in terms of what it does. Because of this your changes aren't "minor". –  Servy Jul 18 '13 at 19:45
    
I don't know lisp, but it looks like you replaced a loop with a simple if. This makes a huge difference IMHO. –  Johannes Kuhn Jul 18 '13 at 19:58
    
@Johannes. "if" is a loop keyword, like unless, but it does the opposite. Here, we collect characters that are not translated (i.e. if assoc returns nil), and when translation occurs we collect the translated character instead. I didn't remove the loop, but I replaced "unless" which did not work.Also, one needs to coerce to 'string in the end to get back a string, not a list of characters. –  Jean-Claude Arbaut Jul 18 '13 at 20:09
    
Lisp is a bit tricky, especially the loop construct, which has many variants :-) –  Jean-Claude Arbaut Jul 18 '13 at 20:10
    
@Bart it kinda makes sense to add a comment but honestly, arbautjc's edit was correct. Even if it was an edit of the answer rather than of the code it's messed up that it was rejected by a moderator who doesn't have experience with the language or any of the tags associated with the question. If there's a way for the mods to suggest why an edit was rejected more specifically, it would kinda help. –  omouse Jul 18 '13 at 20:11
2  
@Servy, I agree it's not so minor, since I changed a not working program into a working one, but the logic is clearly the same. I changed what often occurs as programming mistakes, not the main idea of the algorithm –  Jean-Claude Arbaut Jul 18 '13 at 20:12
2  
@omouse That is not community consensus however. Editing code can have big consequences and that might not always be picked up by the OP when it has unintended ones. The community (and thereby the reviewers) seem to have settled on a "don't touch the code" principle. Leave a comment and let the OP correct it. That you were happy with the edit in this instance does not change that. –  Bart Jul 18 '13 at 20:18
2  
@omouse No, that's not the appropriate use of the site here. You personally may not have a problem with the proposed change on your post, but it doesn't change the fact that site protocol dictates that arbautjc should have posted a comment and allowed you to make the edit (or confirm that it was appropriate for him to do so) instead. In your case you're paying close enough attention and are active enough that you'd notice if you didn't like the edit, but what about a less active user, or a user with so many answers that they don't have time to keep track of lots of people all editing them. –  Servy Jul 18 '13 at 20:18
1  
@Bart, I understand principles, but I think that in a community, rigorous adherence to principles is not always wise... To be honnest, it's probably not the first time I correct a little mistake (yes, only a little one), but it's the first time it's rejected. I suspect usually reviewers understand the language well enough to see why the edit is trivial. –  Jean-Claude Arbaut Jul 18 '13 at 20:22
1  
@arbautjc And I suspect that many of the reviewers of your past edits simply didn't look closely enough to realize you were changing something that you shouldn't have, rather than that they made a judgement call that your category of edits are appropriate. –  Servy Jul 18 '13 at 20:23
1  
Okay I agree with @Servy there's a process for this kind of thing, the only issue is neither of us being made aware of it through the mod's rejected message or through the "how to answer" or "how to edit" docs on the site –  omouse Jul 18 '13 at 20:23
1  
Don't feel too bad though @arbautjc. The edit as such is technically not a bad one. It's just a case of preventing problems caused by those who don't know what they are doing. (You should see some of the suggested code edits...boy....). And you're right, the documentation should be more clear on this issue and the community's consensus. –  Bart Jul 18 '13 at 20:27

2 Answers 2

up vote 2 down vote accepted

When it comes to editing the code in an answer, acceptable edits are edits which fix:

  • obvious typo, for example

    • missing semicolon
    • extra opening parenthesis
    • typo in built-in identifiers (language keyword) name like "Sting" (should be "String")
    • mistaking one package name with another e.g. "python-mysqldb" (Ubuntu package name) and "mysql-python" (PyPI package name)
  • inconsistent indentation/spacing (or simply no indentation at all), for example code like this:

    if ($A) {
      if ($B) {
       foreach (@array) {
      print "something cool";
    }
    }  
    }
    

    take care that you don't edit simply because you don't like the coding style (e.g. brace style) or you don't like the variable names, and also be careful with indentation-sensitive languages (like Python)

If an edit involves changes to the logic of the code or constructs used in the code, it runs a fairly high risk of getting rejected, even if the edit is correct. If the correctness of a code edit isn't immediately obvious because

  • it requires some testing
  • it requires some amount of knowledge in the subject matter
  • and/or reviewers might have doubts whether the original poster of answer would be ok with the edit, or whether the code edit is in line with the poster's original intention

chances are the edit is likely to be rejected.

If you feel that the code in an answer should be modified (and it's not related to typo/indentation/spacing), most of the time it's preferable to communicate/discuss with the answer's poster via comment to explain how and why you think some part of his code should be modified (and also to downvote if you think the code deficiencies significantly degrades the quality of the answer).


Some related Meta discussions regarding code edits in an answer:

Regrettably these subtleties with editing code is not really explained that well in the Help Center. Bart has started a feature request to alleviate this issue. 2

 

1 I think this question is wrongly closed as a duplicate. It addresses code edit in questions and answers, whereas the duplicate only addresses questions. What is acceptable for code edit in question is different from what is acceptable for code edit in answer.

2 I'm not affiliated with Bart and co. I swear

share|improve this answer
1  
Well done on the disclaimer. Nobody will ever know. Wink wink... –  Bart Jul 19 '13 at 10:00

Your edit summary:

removing mistakes

From:

for tr = (assoc char replacements)
          unless (nullp tr) collect char
          else collect (second tr))
    'sequence))

To:

 for tr = (assoc char replacements)
      if (null tr) collect char
      else collect (second tr))
'string))

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

Editing is purely to improve the format of a post, not the content. Spelling, grammar, markup. It is important not to alter the original work of the author. If a mistake is made in the code, the appropriate thing to do is post a comment, not go in and alter the person's code.

When editing it is important to retain the person's 'style' or 'identity'. It is not your post, it is someone elses. If you can improve it for readability or clarity (or for example when English is a second language).. . but not change the content, what the Op has tried, references, examples

Also:

You didn't add the sentence after the points in your quote:

Some common reasons to edit are:

  • to fix grammatical or spelling mistakes
  • to clarify the meaning of a post without changing it
  • to correct minor mistakes or add addendums / updates as the post ages
  • to add related resources or hyperlinks

> Try to make the post substantively better when you edit, not just change a single character. Tiny, trivial edits are discouraged.

Minor mistakes, spelling, grammar, format, are not going through and correcting work, like a teacher, and minor edits are not encouraged.

share|improve this answer

You must log in to answer this question.

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