Recently I flagged a post as possibly being a duplicate of another on stackoverflow. However, this request was rejected as the flagged duplicate was not within the serverfault site.
I found a similar problem when attempting to vote this as a close reason.
Why? This rationale does not seem to be reasonable.
A duplicate with another question in another SE site seems legitmate.
It is legitimate... after it has been migrated to the site that holds the dupe-question. If a question is a duplicate of a question on another site, a true duplicate with identical content from the same user, then a judgment call needs to be made about which site is the question more appropriate for and then move the off-topic question to the right site for dupe-close.
For near-dupes, identical concepts but different words and users, it is far from clear cut. If the topic is one that resides in the overlap between here and there, then it could quite happily live here; there is a preference site-wide to keep questions on their original site unless it needs to move. It could also be a dupe of a ServerFault question. If it is a near-dupe, there is no ServerFault question it could be a dupe of, and is topically gray, drop an Off Topic close vote on it. When the question gets migrated, vote for dupe-close or flag on the target site.
Because of the above, we have no method of handling cross-site duplicates and cross-posts. Cross-posts are the one case where we'd like to have this ability, near-dupes nearly always fail the 'topical' test as well as dupe.
The biggest issue with flagging as a dupe of another site, is that there's nothing we can do about it except close it as off topic. But if it belongs here, and not on SO, then it should have been flagged on SO, not here. But then they can't do anything except close it as off topic either.
But again, I almost never decline a flag unless the user is consistantly pissing me off with crappy flags, and I wouldn't consider your flag a reason for that.
Keep in mind that not all questions that are apparently similar are actual duplicates. It's possible to write superficially identical questions on both SF and SO asking about bottlenecks in SQL server with the intention of soliciting opinions from us about bottlenecks that sysadmins & DBAs see and asking SO about bottlenecks they deal with from a developer's perspective.
Of course whether its possible to ask high quality questions in both sites this way may be another question.
