A few days ago, I asked for my first code review. Now I refactored the code and fixed some bugs, which I want to be reviewed to. But reading in the help center, changing code is a no-go:

Always keep the original code intact!

Reading further, I'm not sure if it's okay to edit it before someone has posted an answer:

Stack Exchange places great importance on answers, so we want to make sure they stay relevant with the question. This is why we ask you to please refrain from modifying the original code based on what has been addressed in answers. By modifying the code in such a way, it'll be harder to determine the merit of the answers, as well as easily seeing the original code that was reviewed.

I can totally understand why I shouldn't modify a question that others have already answered - but what if there are no answers?

At the moment, my question has 30 views, 2 upvotes, but (as I said) no answers and no comments.

Is it okay to edit it or should I create another question with the new code?

  • \$\begingroup\$ The rule is there to prevent you from invalidating existing answers. If there are no answers, feel free to edit. \$\endgroup\$ – Mast Jun 1 '15 at 21:20

You're absolutely right; that header doesn't make sense. This does indeed only apply to questions without answers, which can be edited freely without violating this policy. I'll make this change to that header in the Help Center.

Bearing this in mind, feel free to edit your question until it does get an answer.

  • 2
    \$\begingroup\$ Should we or shouldn't we warn about what happens in case of a race condition? \$\endgroup\$ – Mast Jun 1 '15 at 21:21
  • \$\begingroup\$ @Mast: Other than looking at the precise timestamps, I'm not entirely sure. \$\endgroup\$ – Jamal Jun 1 '15 at 21:22

I'll be even more strict and say it's fine to refine the question as long as no answers and comments are provided.

Sometimes, people do point out potential bugs as a question-comment, probably because it's just one glaring and straightforward point. Therefore, I feel even those shouldn't be invalidated unless the OP and the commenter can somehow agree to delete the invalidated comments afterwards.

In the unlikely event of a race condition, I feel we should let the OP 'win' with the edited question, but this is mainly to encourage newer answers on the non-edited parts. The affected answers can also simply be prefixed with 'This suggestion/advice is coincidentally covered by an edit to the question at the same time'. That will be useful especially if the answer also focused on why the edit is desirable.

  • 2
    \$\begingroup\$ Comments are "second-class citizens" on Stack Exchange sites, and there is a way of flagging comments as "obsolete". Comments are removed much more often than questions and answers. \$\endgroup\$ – Simon Forsberg Jul 22 '15 at 13:24

You must log in to answer this question.

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