Geography 970

March 22, 2010

24 hours in 24 seconds

Filed under: Uncategorized — Jeremy White @ 1:51 pm

Combining the user.location data with the reply-to data (for a full 24 hours) yielded some interesting results.  Twitter activity in the US and Europe is relatively high, but I don’t think that’s surprising to anyone.  However, the amount of activity in Indonesia and Venezuela is interesting.

I started concentrating on the aesthetics more by adjusting the fading circles and gradient lines on the map.  I added an animated time slider on the bottom of the map as a reference, which should help the viewer determine the approximate time of any location.  I also coded the MapNodes script to add the total number of tweets per frame.  Each frame represents two minutes of Twitter activity.

You can see the animation here, but it’s rather large at 18MB.

Advertisements

3 Comments »

  1. Very Cool! The striking feature of this animation is the intra-regional communication for most areas, (Hawaii tweets usually only go to other users on the island chain) and relatively few tweets are long distance. This contradicts some of my assumptions that there would be less regional patterns and more global communication (ideas of openness and lack of barriers or knowledge of distance of communication). Also the use of the scrolling time is an excellent solution to the problem of depiction of time across the entire surface.

    Comment by kjmcgrath — March 22, 2010 @ 2:32 pm

  2. Very nice! The tweets to and from Indonesia are pretty steady throughout the middle of the night. That is wild. While the rest of the world seems to ebb and flow with the sun, Indonesia keeps on Tweeting.

    Comment by Tim Wallace — March 22, 2010 @ 3:29 pm

  3. Nobody else is jazzed for the high quality of the data? Think about it: this is the result of 24 hours of gleaning tweets and 300 thousand RESTful queries. There aren’t any noticeable tweets going into the middle of the ocean. That amazes me, but maybe you all have more faith in VGI. My experiences with the flickr dataset suggests that over 10% of the user supplied locations are crap (e.g. in the middle of lakes, wrong sign on longitudes, thousands of pictures with the exact same location).

    By comparison, the twitter data shown here is superlative.

    Comment by mattmoehr — March 23, 2010 @ 9:55 am


RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Create a free website or blog at WordPress.com.

%d bloggers like this: