tea.mathoverflow.net - Discussion Feed (Is geographical distribution of MO users representative of RW m'ticians)

José Figueroa comments on "Is geographical distribution of MO users representative of RW m'ticians" (12216)

José Figueroa — Thu, 23 Dec 2010 19:48:30 -0800

Thanks, Anton. That's interesting. I did take a look at the public dump and isolated the locations from users.xml, but my script-fu was not good enough to get such statistics. In many cases, it seemed that I had to make a decision as to whether, say, Athens was in Georgia or in Greece.

Anton Geraschenko comments on "Is geographical distribution of MO users representative of RW m'ticians" (12213)

Anton Geraschenko — Thu, 23 Dec 2010 15:59:04 -0800

It's not true that most MO visitors are from the US, but just barely. Here's the analytics data from the last month (296079 visits) for every country generating over 1% of our visits:

Country         Visits      Percentage
United States       127762  43.15
United Kingdom  21078   7.12
Germany     19726   6.66
Canada          15229   5.14
France          13517   4.57
India           7233        2.44
Italy           6246        2.11
Russia          5134        1.73
Australia       4877        1.65
Israel          4633        1.56
Netherlands     4156        1.40
China           4140        1.40
Japan           4114        1.39
Sweden          3892        1.31
Spain           3861        1.30
Switzerland     3572        1.21
Brazil          3094        1.04

(unfortunately, tabs get converted to spaces, so this table looks much uglier for you than it does in my input field)

Scott Morrison comments on "Is geographical distribution of MO users representative of RW m'ticians" (12209)

Scott Morrison — Thu, 23 Dec 2010 14:15:43 -0800

You should look at the public database dumps. This doesn't have IP information, but you might extract some data from users' listed websites. Please report back if you extract anything interesting!

Pete L. Clark comments on "Is geographical distribution of MO users representative of RW m'ticians" (12208)

Pete L. Clark — Thu, 23 Dec 2010 14:02:55 -0800

No, surely not. It is, as you say, strongly biased towards US residents (for partly historical reasons: i.e., the founders are American) and even more strongly towards anglophone mathematicians (which seems inevitable).

I have to say that it would be pretty interesting if someone started up a French language MO. Speaking purely personally, this would be an essentially optimal opportunity to practice my French. (Yes, better than going to France. Although that has certainly been a rewarding experience that I hope to repeat throughout my career, for me the "total immersion" only works to a certain degree. Eventually my colleagues either take pity on me and switch to English, or I get tired and drift off into my own thoughts.)

sds comments on "Is geographical distribution of MO users representative of RW m'ticians" (12206)

sds — Thu, 23 Dec 2010 13:46:53 -0800

Is geographical distribution of MO users representative of distribution of mathematicians as a whole ?

It seems most MO users are from the US.