I thought that this question C# newbie List question was a duplicate (sorry, as a noob meta user I can only post one link, please follow my comment on that question), but other folks don't seem to agree.
If two questions are phrased differently, but describe the same underlying problem (and therefore have the same answers) are they not same question?
Thanks, from a puzzled Binary Worrier :(
Related:
Rule of thumb: If you ask a question similar to another question and it is likely to get the exact same answer, you have yourself a duplicate question.
In your case, take the opposite of that. If the question is similar and it is likely to NOT get the exact same answer, you likely do not have a duplicate.
Other example to consider: duplicating a dead/badly asked question
If you have someone coming and asking a question, but not putting effort in it, not providing feedback or details, this question will most likely get downvoted in the end, and forgotten.
Now what if a new user comes, and has the same problem? He searches for existing questions on the site, and finds a very similar one. However, he doesn't own the question, he can't add details to it, all he could do is post a new answer (leading to the many "i have the same problem" answers, which are sometime useful, with new details, but which shouldn't be answers).
In this case, I think he should ask it as a new question, giving his own details. This new question should contain information about the duplicate (to prevent fast people from closing it after a fast search), something like:
I have searched for duplicate, and found this question (link here). However, this question is abandoned, mostly because the author didn't provide any feedback on his problem.
In this case, the original question should be the one closed for duplicate, and point to this new one. Because we can't let someone "ruin" a problem and prevent others, more motivated, to find a solution.
This particular case of duplicate would help improving the quality of answers on the sites.
If they're phrased very differently, I'd leave them both open. It casts the net wider. Obviously two people seeking the same answer asked two very differently-phrased questions. If you kill one of the questions, you effectively prevent 50% of the potential users from getting their answer in the future simply because they didn't ask the question the way the other 50% of users did.
In instances like this, I will cross-reference each question as an answer, or a comment.
A duplicate is not a duplicate when they are not seeking or asking the exact same thing as something else that already exists.
I don't believe that similar will always be a duplicate, but it can be.
I find it best for authors to highlight similarities between their issue and others, but take it a step further by explaining exactly what they need. This adds distinction between posts...
I think that C# newbie List<Interface> question is a duplicate, but not a duplicate of Why is this cast not possible?.
They are both "why doesn't variance work the way I'd like it to work in C#" duplicates. However, "newbie" is so much simpler that, if it were not a duplicate of many others, it would deserve to stand on its own.
The problem will be finding the other duplicates. The following are from the first page of https://stackoverflow.com/search?q=contravariance+interface+c%23+list+generic. There are ten pages:
BTW, I chose "contravariance" in the above search because I thought there would be fewer occurrences of that word than of "covariance". There were ten pages of each.
One interesting example today (for C++) was somebody asking how to call a constructor manually and build an object on already allocated space. There was a bit of discussion on whether it was a duplicate of "what does placement new do?", since the questions are within easy grammatical transformations of being answers for each other.
