Very much inspired by yesterday's xkds's movie narrative charts, I started another approach in visualizing conversations in Twitter using a time-line mixed with a physics engine.
The chart shows three hours of conversations around the #unibrennt hashtag which is used by students who occupy Austria's largest lecture hall at the University of Vienna to stand up against the devastating status of the country's educational system and where it's heading.
I see lots of room for improvement in this technique, but I just put this together in a few hours to see how it works. Of course it would be very interested to do this with all the data related to the hashtag to analyze communication patterns and to see how the whole process started.
It was done using processing and Rhizome Navigation.
More Images
Video
Owner: Walter Last edited on November 3, 2009 16:39 by Walter / Views: 6642
That's funny that we were both inspired by the same comic! This method of displaying Twitter data is great. Are you using the in-reply-to field to indicate that they were having the conversation?
Also, where did you get data about the hash tag? I thought you could only get data about users.
This would be very useful to know for my Twitter network explorer: http://apps.asterisq.com/mentionmap
Hi! I saw mentionmap the other day, it's great!
I'm using Twitter's search API and parse each tweet with regexp. From that I create a graph representation containing timestamps, users, hashtags and links. So basically it's a graph visualization but the nodes are pinpointed to their corresponding point in time.
Hope this helps ;-)
http://victorhurdugaci.com