After reading a post by user Gopi in the chat room, I wrote a query on Data Explorer which lists the number of questions posted in each month and the number of closed questions among them (although it is not exactly what Gopi asked). I will not try to interpret the data here, but if you find it interesting or if you have something to say about it, let me know.

For the last month in the table, the actual number of questions and closed questions should be larger than the numbers shown in the table because the table is as of the time of the last update of the database.

Questions with no answers, low score, and low activity for a while are deleted automatically. This means that in most cases, closed questions with negative score which are more than 30 days old and closed questions with score 0 which are more than 365 days old will be deleted and not included in the table. Therefore, the actual number and the actual rate of closed questions for more than one month old will be higher than shown in the table.

Warning. I am not familiar with SQL, the language in which this query is written. Therefore, the query may well contain a bug. In case you find a bug, I am happy to learn about it.

Aside. The table shows that there is one question posted in 2008. Unfortunately this is not an evidence for a time machine. This question was migrated from another site, and it was posted in 2008 in the original site. Ideally the query should use the month when the question started to appear on cstheory.stackexchange.com, but I currently ignore this issue because this should affect only the migrated questions and there are not many of them.

  • $\begingroup$ Can't see anything wrong with the query. For comparison, math.SE's close rate is stable between 1% and 4% as opposed to ours (tending up to 20%). $\endgroup$
    – Raphael
    Commented Oct 1, 2011 at 15:37
  • $\begingroup$ @Raphael: Thank you for checking the query. $\endgroup$ Commented Oct 2, 2011 at 18:35
  • $\begingroup$ Thanks @TsuyoshiIto, right after I put my question on the chat I had been wondering if there was a way to find many more DB information but it was too late for editing and I did not want to flood the wall. Fortunately you also answered my other questions. :) $\endgroup$
    – Gopi
    Commented Oct 3, 2011 at 14:22
  • $\begingroup$ @Gopi: Glad to know that this helped you. $\endgroup$ Commented Oct 3, 2011 at 14:55

2 Answers 2


I'd say there's a slightly disturbing trend in the number of questions being asked, especially when you factor in the close rate.

  • $\begingroup$ I agree. I am surprised by this tendency. I partly hope that the query contains some bug. :) $\endgroup$ Commented Sep 30, 2011 at 20:00
  • 3
    $\begingroup$ This spurred me to ask a question about what I am currently working on. $\endgroup$ Commented Sep 30, 2011 at 21:08

This can mean a couple of things:

  1. We have to encourage more researchers to post more questions
  2. We have to encourage more researchers to answer more questions (in order to motivate 1.)
  3. We have to discuss policy/scope
  • 6
    $\begingroup$ I disagree with point 3; our scope has been discussed so many times already, and most people seem to be happy with it. $\endgroup$ Commented Oct 1, 2011 at 12:42
  • 1
    $\begingroup$ I wrote "can mean" very conciously. I remember the discussions but that does not mean their results have to hold infinitely. Our prized scope does not do us any good if our numbers drop to levels that calls "official" attention to matters that should be ours to decide. I am not privy to the intentions of our hosts, but it seems to me that they can not afford to have a weak site so as not to weaken their brand and their position in holding back weakly supported proposals. Fact is, many externals do not like our premise and if we lose our one argument---"it works"---we might lose control. $\endgroup$
    – Raphael
    Commented Oct 1, 2011 at 15:30
  • 1
    $\begingroup$ Therefore, it might be wise to discuss now and maybe struck a compromise lest the decision is made for us in $k$ months. But then, I only wanted to call the point to awareness, not vote to do it. If we can improve our numbers by working harder on 1 and 2, 3 should be unnecessary for a while. So, what about those two? $\endgroup$
    – Raphael
    Commented Oct 1, 2011 at 15:32
  • 4
    $\begingroup$ I agree that we are not getting as many questions as we would like but that doesn't mean cstheory is a weak site, contrary, I think it is quite good. It is good to discuss how we can get more on-topic questions but not with changing the scope to make off-topic questions on-topic just to get more questions. cstheory is useful resource and it is building a good brand between TCS researchers. So yes to 1, no to 3. About 2, some of the questions on the site are really hard and need serious research to get answered. There is no easy way of answering them. $\endgroup$
    – Kaveh
    Commented Oct 1, 2011 at 16:25
  • $\begingroup$ I think what we would like to have is a cstheory of MO quantity and quality. How we can get there is what we should discuss. It would be nice to have some statistics from MO when after 1 year so we can compare and find what we can do to improve situation. $\endgroup$
    – Kaveh
    Commented Oct 1, 2011 at 16:28
  • 5
    $\begingroup$ I also feel that we need to work more on 1. While I don't expect many people to be active all the time, I hope that people are visiting the site regularly, and also hope that they ask questions when needed. For example, I'd like to see more questions like this one: cstheory.stackexchange.com/questions/8449/… - while the question may not require much new thought, it's definitely at grad student or beyond, is a technical question, and got an answer from the right person. $\endgroup$ Commented Oct 1, 2011 at 16:35

You must log in to answer this question.

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