42
\$\begingroup\$

Sometimes I stumble upon follow-up questions, where code is being posted again after having gone through a review (or two).

Examples of possible sets of follow up questions:

Related meta questions:

How should a follow-up question be posted properly?

How to deal with follow-up questions?

\$\endgroup\$
2
  • 2
    \$\begingroup\$ Is is just co-incidence that all these questions are from the same user? Are there other examples? \$\endgroup\$
    – rolfl
    Commented Nov 26, 2013 at 16:17
  • 2
    \$\begingroup\$ @rolfl I believe there are more examples but I haven't kept links to them. When the data explorer works for Code Review, perhaps a query or two can help in finding such questions. \$\endgroup\$
    – Simon Forsberg
    Commented Nov 26, 2013 at 16:23

2 Answers 2

Reset to default
49
\$\begingroup\$

I think that such question should clearly state that it is a follow-up question to a previously asked question.

A good follow-up question should include, in my opinion:

  • Which question it is a follow-up to
  • What changes has been made in the code since last question
  • Why a new review is being asked for

If a question does not include (at least some of) these things, I think that we should close it as a duplicate (or at least give it a down-vote)

There are also some things you should note if you are posting several follow-ups

In addition to the above, each follow-up question should be a good question also by itself.

\$\endgroup\$
3
  • 4
    \$\begingroup\$ Yes I think it's a valid follow up question from same user if he/she learned from the previous review and improved his/her code accordingly. \$\endgroup\$
    – Anirban Nag 'tintinmj'
    Commented Nov 26, 2013 at 21:05
  • 2
    \$\begingroup\$ @doug65536 I am sorry that you feel that way because you haven't got any answers in the last 24 hours on your follow-up question. I believe that the fact that some questions get little attention is not related to whether or not it was a follow-up post. You can try improving the quality of your question, ask in chat if anyone can look at your question, or add a bounty to your question. \$\endgroup\$
    – Simon Forsberg
    Commented May 6, 2016 at 9:10
  • \$\begingroup\$ @SimonForsberg Yeah, my bad. I realized that I had mis-tagged the question as C++14, probably watched by a lot fewer users watching than C++. Nothing in the question was particularly C++14 related. It had a very low "views" count though, which misled me to believe that it was being ignored. \$\endgroup\$
    – doug65536
    Commented May 6, 2016 at 10:51
2
\$\begingroup\$

Everything that Simon wrote. Additionally, you can help readers of the original question find your improved version:

  • Include a link to the original question. This will ensure that each question appears in the other one's Linked Questions list (in the right-hand column, for the current default presentation).
  • Accept one of the answers to the original question. This is a strong hint that you believe that version has been sufficiently reviewed now.

Be prepared for a follow-up review to take longer: the obvious deficiencies having been addressed, reviewers need to spend longer digging further below the surface to find improvements to suggest. And those who reviewed version 1 may intentionally hold back before answering, to allow other reviewers to give different perspectives.

\$\endgroup\$

You must log in to answer this question.

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