The Egyptian Revolution on Twitter

[143mark6275]

they look like sperm.

[MrJayson786]

hyyyyggggyyyy

[happinesson]

cyber revolution is on! Watch out! it is epidemic!

[poletheconvideo]

This is actually bad information visualisation: the most rt'd people are actually pushed to the side in favor of the least rt'd people, only because it creates the 'image' of a 'square'.

Good infoviz stays true to what actually happened, and in this case would have given the most rt'd people centre stage and let the least rt'd drift out of the picture entirely.

[jimiydoorshamelech]

@poletheconvideo

Youtube: Arab Spring Prediction

[vjaypan]

nice video. you ever considered adding a "gravity" modifier, helping those heavy joints to "fall" into the center and push the small conversations to the side? i think that might help to get a clearer view of the growth / connection.

[mikealexaldana]

Me sigue pareciendo biológico el asunto es increible

[jfelixsouza]

1 dislike ---> Mubarak

[Bl4cKeN1nG]

Hope paints such a beautiful picture =)

Looks very decentralized and spontaneous. No one person could have done this.

[pierrelombard]

Actually reminds me of the people converging on Tahrir Square and the building passion...like

[JOSEPHKORSSIA]

What's this? Flock of birds? :)

[JEIWILBER]

Its weird ! ! !

They Must do more of this and broad ! ! !

[marielytgg]

the twitter user on the right must be very popular since he/she got lots of retweets!!! =)thumps up for this graph!

[mikealexaldana]

La comunicación hace que nazcan estos organismos sociales aparentemente de la nada... Estaría padre ver el diagrama ahora (mucho mas disperso supongo)

[renatomassaro]

21st century Art.

[AirRichter]

I do like the visual for the obvious annoying spammer on the right side of the video at around 1:35. And you can tell because very few of the people they tweed'd ended up doing any retweets. They probably have him/her on ignore or something.

[panisson]

@AirRichter The "obvious annoying spammer" is Andy Carvin, and all people connected to him are connected because they actually retweeted him. I would call him very influent, not a spammer.

[AirRichter]

This looks neat and all, but this 'research' involves 3 research teams? Seriously? It's basic social networking theory. This has been the same story on 'viral' emails and other things since forever ago, and even happened with plain old phone calls to a smaller degree (general patterns, when graphed, look the same -- just they were smaller). Twitter just has the benefit of corporate interests who like to see it's name being linked with other popular/emotional public events. (ie, marketing)

[panisson]

@AirRichter This video is only a very small part of the ongoing research in these teams - it took only some hours to create it. For other projects, you can check, for example, the Sociopatterns project, or Truthy at Indiana University, for analysis on information spreading and also analysis of patterns in face-to-face contacts to understand information and viral spreading.

[safwat1990]

I can't understand this from the describition can someone explain simple?

[Dosagu]

@safwat1990 every dot is a twitt every dot than conects is a retwitt, yo can see what happends during the time.

[panisson]

@Dosagu Every dot is a Twitter user. Each time a user retweets other user's tweet using the hashtag #jan25, a link appears in the network.

[urbanjahvier]

Amazing! 

[FoodOrFail]

Cool. Could you please discuss the formula used (in general terms)?

[jnts]

Wow!!!! Fascinating!!!! Fastastic this video!!!!

[cultware]

(sorry damn double clicking!)

[cultware]

3.38 is the same length as David Bowie's Heroes. It makes a great soundtrack

[fredericsidler]

I love data visualization and this one clearly shows the power of Twitter. Great work André. Which country next? Which hashtag? Which revolution?

[GephiOrg]

Nodes are twitter users, and links appear between the nodes A and B when B @zappullae retweeted a message of A containing the hashtag #jan25. The location is done with a force-directed algorithm, which works by attracting two nodes with a link, and repulsing the others.

@MehtaMezhan yes it's a screencast of Gephi!

@Squeeonline yes there are many questions, the fact is that this data was captured nearly by surprise. I guess that a deeper analysis will be considered in the near future.

[zappullae]

@GephiOrg thanks. its very interesting representation

[zappullae]

Very cool! Can you explain a little more about how the data is represented.

Is each Node a tweet and the line segments a re-tweet of the connecting node?

Are the segments directional?

How are the locations of the nodes represented spatially, geo location, tweet time, etc specific?

[MehtaMezhan]

Interesting, but the framerate/quality drops drastically in the later (most interesting) portions. Try doing a pre-render at native 720p and uploading afterwards; it looks like a screencap video.

Some Creative Commons music wouldn't hurt either! ;)

[Squeeonline]

Wonder how many views will come from /. ?

Also is it fair to use only the one hashtag? Is it a fair representation? Very interesting video anyway.