I posted a question to SO, and it was closed as a duplicate, I then posted several comments pointing out that it was NOT a duplicate, that all of the answers to the supposed duplicate were not relevant to my question, and it was reopened. One person answered (with a subjective answer), and the question was closed as not constructive.

Where should you argue that a question was closed incorrectly? Just in the comments? Those quickly get hidden due to the number of comments. The accepted answer to this meta question sorta implies that posting it to meta is the right way, but that seems wrong to me (not sure why, just feels off). Or maybe there's a Meta tag that should be used.

EDIT: Note that while I'd of course like it if the question was reopened, this is intended as a serious question for future reference if and when my questions are closed.

  • 3
    You've come to the right place. – jscs Jan 12 '12 at 18:22
  • 1
    If the argument cannot be (or isn't) resolved in comments, posting on meta is the right thing to do. – Adam Lear Jan 12 '12 at 18:23
  • 3
    Can you explain more about how the answer posted doesn't actually answer the question? – Charles Jan 12 '12 at 18:24
  • @RobertHarvey: Three reason's. (1) The answer is in a comment so I can't accept it, (2) it's only a partial answer and I'm hoping someone can come up with the other part, and (3) closed questions are more likely to be skipped. – jmoreno Jan 12 '12 at 18:58
  • @Charles: MSDN, which was referenced in the (at the time) single "answer", says just that they don't, but doesn't give a reason as to why. Daniel's comment OTOH, for why it won't work for nullables, is a technical reason, but I don't think the same logic applies to sqltypes. If it was an answer, and there was no one else that wrote an answer that included a reason for SQLTypes, I'd accept it as the best answer. – jmoreno Jan 12 '12 at 19:35
  • @jmoreno, maybe I don't understand, but the answer as given seems extremely simple, logical and concise. It clearly and obviously explains that SQL null is not a thing that can be represented as a CLR null, which is a truth. In fact, I can't think of too many other languages that have anything similar to SQL null. Can you explain more why this answer is not satisfactory? Maybe I just don't understand the actual intent of your question well enough. – Charles Jan 12 '12 at 20:00
  • @charles, it clearly says that they are not equivalent, it doesn't explain the difficulty in converting from db null to clr null. It doesn't explain that this would have to be a runtime conversion because you are never dealing with a variable of type DBNull, but instead an Object that contains a DBNull. Basically I'm trying to understand the technical constraints that lead to the implementation. – jmoreno Jan 13 '12 at 6:31

Stack Overflow is a place where people can come to get help for their programming problems. But not all questions are equal.

While questions and answers may be useful to the person asking the question, they are not always useful to everyone else visiting the site. Consequently, some questions remain open and stay on the site, while others get closed and are eventually deleted.

This, of course, is the place to come for close appeals. Comments on the main site are really not a good place to argue for a question reopening; they cause noise, and are generally not heeded.

  • I agree with the noise, that's really why I was asking this question, well that and I don't think arguing in the comments is very effective as practically no one will see it. Do you just post a question to Meta with a link to your question? – jmoreno Jan 12 '12 at 18:55
  • Yes, that's what you would do. I reopened your question, and cleaned it up a bit, after discussing it with the original mod who closed it. – Robert Harvey Jan 12 '12 at 18:56

You must log in to answer this question.

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