What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 130 Stack Exchange communities.

So I found a useful post on SO earlier today but found that the code given is no longer correct

http://stackoverflow.com/suggested-edits/122517

However, it was rejected as "This edit is incorrect or an attempt to reply to or comment on the existing post"

The edit was correct, so much so that trying to use the approved answerer's code would produce a compile error. I feel that pointing it out in a comment is not so helpful as actually correcting the code, especially when there are already nested comments.

Perhaps people voting on an edit should be required to have a certain experience level in the tagged topic in hand to vote on programatic changes? I know I would much prefer to have a working solution to a problem as the accepted answer as opposed to broken code and a fix 8 comments in...

share|improve this question
2  
Submitting an answer doesn't require any approval... –  John Oct 19 '11 at 19:48
    
On the plus side you're only 35 points away from the "edit questions and answers" privilege so it won't matter to you much pretty soon –  Some Helpful Commenter Oct 19 '11 at 19:50

2 Answers 2

up vote 6 down vote accepted

That edit should be ok for an answer (while it wouldn't be for a question). I've done small code fixes to answers before where it needed to be done.

If you resubmit it, I'll vote to approve.

share|improve this answer
    
I think I'd reject it, never know if that difference was expected/intended/"how things were done back then". Given how prolific the OP was, a comment asking about it would have been more appropriate. –  user7116 Oct 20 '11 at 0:31
    
Ok I've resubmitted that edit. @sixlettervariables: The whole point of my post is that leaving it "how it was done back then" isn't particularly useful if it is deprecated code that won't work anymore. If something is wrong the user's rep level shouldn't even enter the equation –  davbryn Oct 20 '11 at 8:29
    
@davbryn, it got rejected again, but I went in and directly edited it in. I'm putting faith in you that the edit is correct, since it's not my niche. Congratulations on hitting 2k, now you can directly edit. Use your powers wisely. –  Lance Roberts Oct 20 '11 at 14:24
    
Thanks Lance - The edit is definitely correct, I was implementing the code when I discovered the deprecation! –  davbryn Oct 20 '11 at 14:47
    
@davbryn: my point was if the post is specific to a version (or intended to be) you may wish to check with the OP via a comment before editing it. If it were a simple typo, no problem. Not everybody codes on the latest and greatest either. –  user7116 Oct 20 '11 at 21:21

I'm guessing the reviewers in question aren't familiar enough with NSTimer (or possibly even Objective-C) to determine if that is indeed correct or altering the answer and behaviour of the code.

It's pretty hard to work that out quickly for things you're not familiar with. For edits like that (as opposed to grammatical/spelling/formatting edits) I would normally pass over them and defer to someone who (hopefully) has more experience in the tags.

The obvious "fix" that springs to mind would be requiring a minimum amount of rep in a tag in order to approve edits, but the problem there seems to be that it would significantly reduce the pool of people eligible to review (which is bad for the simpler to judge edits).

To work around that there could be an option when submitting an edit to require reviewers have tag relevant experience. I don't think that would work well in practice though because it might end up being abused. One solution might be to make that automatic based on editing the content of code blocks - you'd want to avoid whitespace counting for that, although that in itself might be problematic for at least Python.

The other option (beyond simply resubmitting and hoping you get different reviewers) would be a flag option on the suggested edit page itself. Under the hood this could be just window dressing on the existing flagging mechanism and raise an "other" flag on the answer in question pre-filled with a link to the edit. There's a penalty there for abuse (loss of flag weight) and "oversight" from higher up the chain. Alternatively the flag this suggested edit might move the edit back into another "require tag experts to review" queue.

Both of those seem like a lot of work to implement though over the existing options (flag answer or resubmit edit) and it's quite an edge case too so I doubt they'll happen.

share|improve this answer

You must log in to answer this question.

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