# Is it normal for questions to be closed by moderators in private beta?

I've just noticed that this question and this question were closed by @Rebecca Chernoff. It seems both question already had close votes, which is why I think they ended up getting closed.

My guess is that the SE Network has decided to do this to help communities in private beta to get enough close votes on questions that should be closed? If so, is this policy documented anywhere?

Yes, it is normal. And no, it's not so much to "help them get enough close votes" - it's to make sure that any problems with questions in the early beta period get discussed.

Remember, when this site exits private beta, the questions that have been asked, and the discussions here on Meta are what set the tone and define the scope for new users entering the site.

Questions that are borderline on-topic, poorly-asked, unclear, or suffering from other issues must go through this process in order for the site to launch with clear guidelines on what questions should be asked and how they should be presented.

In the specific cases you cite, both of those questions had had concerns raised that had not been sufficiently addressed by the authors or others. So they were closed in order to prompt either discussion or revisions. There are other outstanding close votes on questions that have been thoroughly discussed and addressed; whether or not these are closed is up to the rest of the community.

• Ok, fair enough. I understand the reason and agree with the goal; I do fear that closing these questions may incite resistance rather than discussion (particularly if the goals you mention are not communicated well or not understood by the other party). I can't think of a clearly better alternative though. – Alex ten Brink Mar 12 '12 at 17:24
• If they do incite resistance, that needs to be dealt with early as well. One of our failings in the past has been poor communication of what closing is meant to achieve: it's not a death sentence, it's a way to hold a question while issues are resolved. When I see folks talking about whether or not a question "deserves" to be closed, I start to worry: if they can't correct the problem themselves, and the author can't or refuses to, it needs to be closed until he does - not as a "punishment", but to prevent it from causing problems for all concerned. – Shog9 Mar 12 '12 at 17:28

I don't know about the type checking question (148), but the SQL question (135) should not have been closed. One user expressed they did not like the question whereas at least eight liked it enough to upvote. Kaveh gave sufficient reason why his question should remain open, but this explanation was ignored.

I wonder: do the community mods who monitor here have expertise in computer science? If not, I wonder how they can judge the value of a question. Closing anything with a close vote on it is a poor strategy.

I am not sure closing is a good method to prompt OPs to adapt their question; closing seems harsh, and we know that few users read FAQs right away. Maybe a private message or a comment would be better suited to ask users to adapt their questions in a friendly way?

• I think your first paragraph is not an answer to my question - if it concerns you, maybe you should begin a new meta question on that topic. I also think there are more constructive ways of phrasing your suggestion ('a private message or comment may be a better way of prompting the OP to adapt their question'). Maybe that should be another meta question(/suggestion) as well? I think Shog9 answered my question quite completely. – Alex ten Brink Mar 12 '12 at 17:36
• It takes so little effort to edit... If folks aren't understanding your question, then clarify it. Preferring to argue with them in comments, or waiting for someone to post an answer and then explaining why they've wasted their time on a misunderstanding... Is simply rude, and sets a poor example for others. – Shog9 Mar 12 '12 at 17:37
• As the two questions are the examples that prompted your question, I think it is fair to evaluate the policy @Shog9 states for them, as it is to discuss the policy itself. – Raphael Mar 12 '12 at 17:40
• @Shog9 In the case of 135, the problem was not the question, imho. – Raphael Mar 12 '12 at 17:42
• @Shog9 If someone asks a question on Stack Overflow about a specific language or library and someone who doesn't know that language/library complains that he doesn't understand the question, would the author of the question be expected to explain the basics of the language/library as part of his question? – sepp2k Mar 12 '12 at 17:52
• @sepp2k: if confusion arose out of a misunderstanding as to what the question was asking for, the author is expected to clarify it, yes. Note the distinction, and please read the comment thread on 135: no one is asking for a tutorial on descriptive complexity, merely a clarification as to what the author is really asking for (which is already provided in the comments). Since both you, Raphael, and Alex are misunderstanding this, I suspect I need to address it in another post here on Meta. – Shog9 Mar 12 '12 at 17:59
• @sepp2k, Raphael, I got sidetracked after encountering a much larger discussion, but you might be interested in: meta.cs.stackexchange.com/questions/78/… - I do in the end touch on what I suspect may be a difference in philosophy here. – Shog9 Mar 12 '12 at 20:20

It is normal for the Stack Exchange team to intervene when they feel that the community is behaving in a way that is detrimental to its survival. Their actions can be reverted by the community, preferably after discussion (the kind that is happening here — prompted by the closures, as it happens).

Speaking of these two closures in particular, Shog9's answer explains why they happened. I'll give my opinion on the two questions in question:

• Extension of SQL capturing $\mathsf{P}$: I think this is a reasonable question; it is outside my area of expertise, but I do not doubt that it makes sense as is to experts. However, the very existence of the comment thread shows that the question needs clarification (if only to avoid attracting wrong answers from readers who misunderstand the question because they lack the necessary prerequisites). I'm not familiar enough with the material to edit myself, but I would vote to reopen if a short explanation or a pointer to what “capture” means in this context and what qualifies as a query language.
• Type-checking algorithms: As I wrote in a comment, I find this question far too broad; type-checking in the three languages cited as examples (C++, Java, Scala) is very different. I would reopen the question if it either focused on one of these languages, or gave some overview of the kind of language that the asker wants to typecheck. I cannot improve the question myself since I don't know what the asker is after.

Please remember that closing a question means that it is not suitable in its present form. A question can be reopened if it is edited, and this is what I expect to happen in both cases here. The normal workflow is to vote to close and later reopen, especially in the private beta when everyone can vote and questions are subject to especially high scrutiny and high standards. I recommend reading:

• Regarding my question, I don't see how existence of a comment thread justifies closing or anything for that matter. J.D. posted a number of comments that I replied to explaining that his is not understanding the question. The question is tagged descriptive-complexity and finite-model-theory and I have linked to a theorem in a book on this topic, this should be sufficient I think, if someone is not familiar with a topic (e.g. doesn't understand what capturing a complexity class means) they should act conservatively. Considering the scope of the site, it is going to be quite frustrating if – Kaveh Mar 12 '12 at 20:33
• people who have no clue what is being asked and what the tags mean act as J.D. did. I cannot explain every time I ask a question every elementary point (e.g. a language capturing P cannot be Turing-complete) to make the question sensible to every reader. – Kaveh Mar 12 '12 at 20:36
• @Kaveh When you ask a question in a public forum, you need to manage all readers to some extent, not just the ones who might be able to answer the question. Even if you don't care to or cannot make your question more accessible, you need to do your part to avoid having your question misinterpreted. I've edited your question and voted to reopen accordingly; since I am not familiar with this topic, I hope I haven't introduced any mistake, please review. – Gilles 'SO- stop being evil' Mar 12 '12 at 20:58
• (ps: thanks, I will check it). – Kaveh Mar 12 '12 at 21:43
• I understand that and I am in favor of making questions as accessible as possible, but it is impractical to explain everything in a question, so the question is: where should we draw the line? Should I explain to someone without basic knowledge of complexity that an elementary fact about complexity theory that language capturing P cannot be Turing-complete? Should I explain every standard terminology I use in my question? I hope not. For me, the comments were something like: "why don't you define a class for it in C?". Should I include in my question that C doesn't support classes? – Kaveh Mar 12 '12 at 21:44
• @Kaveh No, and I'll quote a comment thread on a question of mine. Paraphrasing: commenter: “Can you clarify […]? I can't follow because I don't know Coq.” Me: edit to clarify some points, and “I've tried, is that better now? Honestly, this question is pretty specific to Coq.” Commenter: “Much better, thanks.” In both your and my case, I think the question was originally ok, but the reactions in comments showed that an edit was needed. – Gilles 'SO- stop being evil' Mar 12 '12 at 21:47
• @Kaveh Also, some advice from a moderator to a moderator: sometimes, a simple edit that you don't feel is necessary but isn't harmful either is all it takes to make everybody happy. – Gilles 'SO- stop being evil' Mar 12 '12 at 21:49
• Point taken, thanks. :) ps: I edited my question and added some motivation to it as Shog9 suggested. – Kaveh Mar 12 '12 at 21:57