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.

As a edit moderator (2k rep), I now see many many many edits that are actually new users answering the question. This most often happens where a code modification occurs.

Many of these answers look good, and probably solve the question — so it seems a shame to destroy these answers.

Can we have a new moderation response that more explicitly states (to the editor / answerer):

Your edit needs to be submitted as an answer to the question. Modifying the original question creates confusion, since the question is no longer valid / needed.

The typical pattern is

  • modifies the code
  • uses the edit-comment to explain how their edit answers the question

A super-feature would be to allow such moderation to auto-convert the edit into an answer. This would be the best UX practice, IMO, since this is what the user should do, and we can make them love the site even more this way, encouraging more interaction.

So, we would

  • check to see if the edit is code-only
  • create a modifies block of code and set it to the body of a new answer
  • use the edit-comment as a string of unformatted, sentence-case, text that is above the code block.
share|improve this question
    
Certainly that would be nice, of course you can include a custom reject reason to this effect but a canned version would be nice (assuming anyone reads those anyway) –  Richard Tingle Oct 8 '13 at 13:13
    
Not sure how many of these I see, to be honest. –  Undo Oct 8 '13 at 13:13
3  
Do you get a notification in your inbox if your suggested edit is rejected? –  Bill the Lizard Oct 8 '13 at 13:14
1  
@BilltheLizard: I don't think so, never had a notification to such effect. –  Qantas 94 Heavy Oct 8 '13 at 13:18
4  
@BilltheLizard Nope. –  Mansfield Oct 8 '13 at 13:19
2  
Thanks. That might be a simple enough solution then. –  Bill the Lizard Oct 8 '13 at 13:21
3  
Personally I don't think auto-creating an answer is the way we want to go here. –  Aaron Bertrand Oct 8 '13 at 13:26
    
@Undo I saw 3 this morning, and after seeing 1-3 per day on most days, I decided I was fed-up and asked here on Meta ;) –  New Alexandria Oct 8 '13 at 13:28
    
@AaronBertrand why not? The user wrote an answer - it just wasn't in the correct format. We should help them out and save their effort –  New Alexandria Oct 8 '13 at 13:29
3  
Because maybe it's an answer but they want to provide context to help guide a better answer. Or maybe they don't want to post an answer. You shouldn't force them to do so. Reject the edit, and they can decide to re-word the edit, create their own answer, or leave it alone. –  Aaron Bertrand Oct 8 '13 at 13:34
    
@AaronBertrand based on what you've written, I think I've unclear thus far: most of these edits I see & am talking about here are code changes. OP posts code in the Q, someone edits the code to make it work correctly. Should be an Answer, no? –  New Alexandria Oct 8 '13 at 13:46
1  
You didn't say code once in the question, never mind imply that the edits only affected code. In any case, I still think the edit should be rejected (if you are familiar with the language and are absolutely 100% that the edit solves the problem, and isn't an attempt to clarify), but do not agree that creating an answer on someone else's behalf is a "super-feature" or even a remotely useful on. –  Aaron Bertrand Oct 8 '13 at 14:19
    
Your example is a case of the question asker editing an answer to make it better -- it's not creating a new answer. –  dcaswell Oct 8 '13 at 14:36
    
@dcaswell agreed, I didn't have easy access to an older example, and clacked off thinking this was close enough –  New Alexandria Oct 8 '13 at 14:42

You must log in to answer this question.

Browse other questions tagged .