What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 127 Stack Exchange communities.

Check this scenario,

  • A user posts an answer to a question, but that has some issue, as it'll not work completely
  • Another user edits that answer, fix that issue, and comment it as edited for improvement.

Will this edit lie under radical change?

Although I have just approved a Stack Overflow question, http://stackoverflow.com/review/suggested-edits/1528649, due to some humanitarian reasons, I hope I didn't make a sin.

share|improve this question
2  
Which edit is that? Give people this edits' link.. –  Soner Gönül Feb 15 '13 at 14:54
    
Just for informational purposes as it doesn't change how it should be answered, but the individual who suggested the edit the OP of the question. –  psubsee2003 Feb 15 '13 at 14:59
    
@psubsee2003 Sorry! please elaborate.. Specially that OP part. –  rptwsthi Feb 15 '13 at 15:02
1  
The individual who submitted the edit you are asking about is the same person who asked the question. Probably tried it, didn't work, but it helped him enough that he was able to figure out what was wrong and wanted to edit the answer so it was correct. –  psubsee2003 Feb 15 '13 at 15:03
    
Ohk, that was analytical! :) Now comes, what if......? –  rptwsthi Feb 15 '13 at 15:06
    
I would not call that a radical change, as it merely fixed a syntax error with the answer. But the "Edit" note that got added wasn't necessary, as it was already in the Revision History comment. I would have probably either approved it since it wasn't anything major, or hit "Improve" and removed that extra edit note. (depends on how lazy I'm feeling). I would not have rejected it though, as the syntax correction improved the answer :) –  Rachel Feb 15 '13 at 15:09
add comment

2 Answers

I'd call that a radical change. It's a change to code, not a minor syntax error, like a missing paren or semicolon at the end of a line. Regex is tricky and you may have introduced some new bug. Changing a couple of characters in regex can be equivalent to adding an entire new if() or for() in normal code.

I only change badly formatted code or the most obvious of syntax errors. Just leave the code alone and leave a comment for the author.

share|improve this answer
add comment

Is this a radical change? Probably not.

A radical change to me is something that is someone coming in and removing virtually everything from the post and reposting their own answer in their own words. (It's similar to vandalism, except a vandalized post usually doesn't attempt to provide a new answer, it usually just adds nonsense or unnecessary commentary). Like everything else, it is subject to your own interpretation.

My interpretation of "Radical Change" does probably fall on the extreme side of the spectrum. But a more liberal interpretation would probably be something that changes the answer, rather than fixing a small typo. This itself can be a gray area as a minor edit could make a significant change to the answer, so you sometimes have to make your own call as to if it is "Radical" or not.

It also sounds like you were asking how you should handle this. In general, editing code in answers to change the context of an answer is a big gray/grey area. There are 2 schools of thought:

  • Do not change an answer. You should only edit to fix code markup and grammar mistakes;
  • If the code is wrong, and especially if it is an accepted answer, it needs fixed so we are not promoting incorrect answers on the site.

I don't think either extreme is always correct as it depends on the situation and your own knowledge of the technologies involved.

When I am reviewing an edit that is changing code in an answer or changing wording that would change the context of the answer i always ask myself:

  1. Am I familiar with the language and the concepts in the question and the answer?
  2. If so, is the original code actually wrong?
  3. And if it is wrong is the edit accurate?

If it meets all 3 criteria, then go ahead and approve. If it fails any of the 3 tests, then I usually reject or skip.

If it fails #1 and if the editor provides a solid reason in the comment for making the edit, then I'll usually skip and let someone who is more versed to judge the technical merits of the question.

If it fails #2 or #3, then I'll usually reject as an "Invalid Edit".

In the edit you referenced, I do not see it as extreme so I definately would not have called it a "Radical Edit", and I probably would have skipped it since my RegEx is rusty and it would have taken me too long to actually determine if the edit is correct or not, and if it needed changed to begin with. I would not have rejected though in this case because it was the OP of the question that actually made the edit and my logic would have been that the OP tried it, saw it was close but didn't quite work for him, and it needed tweaked slightly.

However, if I was versed enough to approve it, I probably would have selected Improve instead and removed the "EDIT" section that was added. The revision history should be enough that you don't need to put an edit header for every little change.

share|improve this answer
    
A radical change to me is something that is someone coming in and removing virtually everything from the post and reposting their own answer in their own words. That's an extreme example (along with flipping the meaning of the post). Some less extreme example includes changing the code in the question (big or small), and big changes in the code for answer (for answer, this is subjective and needs to be judge case-by-case). –  nhahtdh Feb 15 '13 at 15:32
    
@nhahtdh I see your point, but like everything else it is open to interpretation... radical means extreme, so I've come to the conclusion that it should be interpreted as such. But I will add a qualifying statement –  psubsee2003 Feb 15 '13 at 15:50
add comment

You must log in to answer this question.

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