twitter is a live broadcast medium.

the consumers of the content contribute in real time. aggregations form spontaneously in the data. tweeterbrowser is a data visualization tool designed to diagram these aggregations in the data, spontaneous or not.

the following diagram is a detail view of a tweeterbrowser diagram depicting the relationships between 40 randomly chosen twitter accounts that sent messages containing the text #twitterfilms.

each red dot represents a twitter account. a green line between two twitter accounts indicates that one twitter account is following another. therefore the diagram indicates who knows who on the twitter social network.

the messages used to generate this diagram were gathered between 12:55 PM EST and 1:00 PM EST on Sunday May 24 2009, while #twitterfilms was the top trending topic. 

so what were these messages containing the text #twitterfilms, and who were the users sending these messages?

a quick reading of some of these messages reveals that people were thinking up titles of existing movies and trying to find clever ways of inserting twitter semantics, taxonomy and syntax into the titles. 

the results are as funny and as poignant as the films being given this treatment.

this diagram is the full view of the detail from above. it shows all 4170 users loaded into the diagram in association with our search term.

this diagram was generated using the tweeterbrowser. the tweeterbrowser is a cross-platform data visualization tool that can be used to study participant behaviour on the twitter social network.

tweeterbrowser works by querying the twitter infrastructure in real time to gather the data used to generate the visualizations.

the application has been tested on Windows XP, Vista, Mac OS X 10.5 and Ubuntu Linux. 

this tool works best on a fast computer with at least 1 gigabyte of RAM and a fast connection to the internet.

install the twitbrowser to generate your own visualizations.

http://www.youtube.com/watch?v=zDg4heP_TV0