12
\$\begingroup\$

On Stack Overflow itself it makes perfect sense to link to similar questions and say that the user might get better results from searching those results first and can likely find their problem already solved.

However, for Code Review, it isn't the intention that people need help with their specific implementation, and even if their goal is similar to others it's valuable to get someone reading over their code. It's a slightly mixed signal to think that the intent of the site is for other people to review your work for feedback but then when typing up a question to ask you're told that you maybe shouldn't bother because other people got feedback on their solutions.

I think it's valuable to have the links. I just wonder if the phrasing of the text might throw off the beginners that need help the most.

\$\endgroup\$
  • 2
    \$\begingroup\$ One thing does not have to exclude the other. I think it is helpful to say "this code smell that you have here is very similar to [this other thing](link)" \$\endgroup\$ – Simon Forsberg May 5 '15 at 10:31
  • 1
    \$\begingroup\$ But isn't it very possible for one of those links to "already have your answer"? How many "good" implementations of FizzBuzz in [langX] are there? 3, maybe 4 at the most? I know it's a silly example, but I think it illustrates the point. \$\endgroup\$ – RubberDuck May 5 '15 at 10:33
  • 4
    \$\begingroup\$ @RubberDuck A better example would be Project Euler and other programming-challenge questions. \$\endgroup\$ – 200_success May 5 '15 at 16:18
2
\$\begingroup\$

In one sentence: I think on Code Review labeling something as a duplicate should be done sparingly.

I think this touches on a greater "problem," however, in that a strict interpretation of the StackExchange ethos would say that posts must ask a question, and in response, one answer must rise to the top as "the" answer. I will add that I don't really think it's a problem per se as it is dealt with and discussed well here.

You run into trouble when you start to consider the definition of a duplicate question in light of that framework, or ask, when does my question differ sufficiently from another question? In our day-to-day programming, we are generally dealing with problems that have already been solved innumerable times before (and probably by people smarter than us). Code reviews as we know them don't necessarily require a convergent process to arrive at "the answer" because for the most part it's acceptable to simply say "hey take a look at this code I just wrote" without having to guarantee its uniqueness.

I think that the concept of duplicates on Code review is somewhat of a gray area, but it should ultimately stay gray unless we start experiencing a real need to remove ambiguity.

\$\endgroup\$
  • \$\begingroup\$ I'd be curious how many questions actually get closed as duplicates on this site. \$\endgroup\$ – SuperBiasedMan Jul 22 '15 at 15:33
3
\$\begingroup\$

I suggest the wording

You may read the following reviews on code that solves the same problem as yours.

\$\endgroup\$

You must log in to answer this question.

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