Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 158 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

Now that StackOverflow is a relatively mature site (approaching 300,000 questions), I think it would be quite interesting to take a look at its growth model - in other words, how the total number of questions has varied over time. What might be even more curious is the growth model for some of the more widespread tags such as c# and java.

The model of growth for the English Wikipedia is widely considered to be approximated well by a logistic function, though it has largely been near-exponential so far.

Logistic Growth Model ~ 1/(1+e^-x)

Of course, the nature of StackOverflow is quite different to that of Wikipedia. Although it was designed to in part be a wiki site, it is not often used in a very collaborative manner. It may well be logistic, exponential, quadratic, or one of a number of others...

As far as an explanation goes, here are a few points to consider:

  • The number of users has increased rapidly since the beta period. More users mean more questions. I wouldn't want to guess what model user growth follows, but it's certainly much steeper than linear.
  • Much of the answering (in particular the accepted answers) would seem to be done be a "core group" of the top 100 or so members (Stackers?).
  • Many of what may be called common, basic, or "newbie" questions on StackOverflow have already been answered thoroughly (in many cases a number of times). This discourages the creation of new ones, or at the least they get closed soon after appearance.

So, has anyone yet analysed the statistics for StackOverflow? Is the raw data even available so that we might do so ourselves? Any info in this respect would be worth looking at in my opinion.

share|improve this question
Noldorin, you (normally) can associate your account with SO and gain a 100 rep boost here on Meta. In your case no related accounts are found (user page -> tab "account"). Looks like you should email the team to fix that. – Ladybug Killer Sep 7 '09 at 8:46
Thanks for that, Mehrdad. – Noldorin Sep 7 '09 at 8:46
No problem. :) – xmm0 Sep 7 '09 at 8:47
@John Smithers: Yeah, seems I did it for my and accounts, but not the meta one. Cheers for pointing that out. – Noldorin Sep 7 '09 at 8:49
logistic? doesn't that have to do with bringing-in supplies and people.. did you mean logarithmic? – warren Sep 7 '09 at 11:56
@warren: Not at all. The logarithmic function is the inverse exponential, whereas the logistic function is 1/(1+e^-t). See the link in the question. – Noldorin Sep 7 '09 at 12:18
up vote 4 down vote accepted

To answer the question regarding the RAW data. You have few options:

  • The monthly data dump which is available from here. It is most likely the best option for this type of statistics
  • StatOverflow hosts a online version of the data dump which can be analyzed at will.

There is also the Stackoverflow data dump sites? question that contains a list of other places the data dump is available. Also the data-dump tag will provide you with endless information.

share|improve this answer
Thanks, Diago. Might get around to analysing this at some point. – Noldorin Sep 12 '09 at 13:09

You must log in to answer this question.

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