2

I have an review in my queue where the only change was to add a block of code as an example. My initial reaction was to reject it but none of the rejection reasons seemed appropriate.

In general, is this something that is considered allowable? If not what reason should be used to reject it?

Gnat pointed to the specific question in the comments but I kind of want to talk about a more general case. The reason I considered choosing is "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."

What made me stop is that that reason suggests that it's OK to make drastic changes (such as adding code examples) as long as they are consistent with the author's intent. I didn't yet take the time to evaluate whether that code is consistent but assume for a second that they are. If, hypothetically, the additional code was a perfect example of what the author suggested, would it be OK? In other words is there line where a change is can be considered too drastic?

  • are you talking about this review? programmers.stackexchange.com/review/suggested-edits/148706 – gnat Aug 26 '16 at 17:49
  • @gnat That's the one, yes. – JimmyJames Aug 26 '16 at 18:05
  • @gnat i see you rejected it with the 'deviates' reason. That's where I was unsure. I'll add some more detail to the question. – JimmyJames Aug 26 '16 at 18:08
  • in the hindsight reason picked by second reviewer looks like a better fit to me "...should have been written as a comment or an answer." – gnat Aug 26 '16 at 18:10
  • @gnat I also considered that but blocks of code don't look too hot as comments. Should the appropriate solution be to add another answer and put in a blurb like "to add to blah blah's answer..." – JimmyJames Aug 26 '16 at 18:14
3

In general, moderation should not require technical knowledge of the subject matter.

A code change is almost always a change in the poster's intent, and reviewers shouldn't have to know anything about what the code does (or whether it works or not) to make that evaluation.

Therefore, code examples added to someone else's answer are usually considered a radical change, and can be summarily rejected on that basis.

  • I like this answer. I wonder if this could be added somehow as an option for rejection reason. Not sure what the language would be though. – JimmyJames Sep 8 '16 at 17:31
  • 1
    "clearly conflicts with author's intent" – Robert Harvey Sep 8 '16 at 17:51
  • But that's the issue. It doesn't 'clearly conflict'. It could be an example that is extremely well-aligned with the author's point. In that situation I think the editor would be confused by the response and not understand what they did wrong. – JimmyJames Sep 8 '16 at 17:55
  • But to evaluate that, you would have to know the technical details of the edit, a condition which moderation specifically disclaims. – Robert Harvey Sep 9 '16 at 22:50
  • Right which is why saying it's "clearly conflicting" is an incorrect desciption. The problem is the exact opposite i.e. it's "unclear" whether the addition conflicts with the author's intent or not. – JimmyJames Sep 12 '16 at 16:14
1

I think for such cases it is a good idea to ask the original author if the code example matches his intent and let him make the decision. For the particular case, it seems the editor actually did that, but the edit was so quickly rejected by @gnat that the OP did not even have a chance for this.

Of course, when the OP does not show up here on SE within a reasonable time, or does not react at all, (to me, 72 hours looks quite reasonable), such an edit should probably be rollbacked.

You must log in to answer this question.

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