The self-organizing student protest at University of Vienna

On October 22, hundreds of protesting students occupied the Audi Max lecture room at the University of Vienna in order to create more awareness for their situation and their demands for a better university. Actually, the quality of the studies decreased when Austrian goverment de facto waived the tuition fees creating a flood of national and international students into particular studies. Since that date, the students still occupy the lecture room.
While I am very sceptic that their demands for a high quality study with a low student-to-professor ratio together with an unlimited offering of studies at no charge can be ever fulfilled within a reasonable budget, I am very impressed by the organizational structure of the protesters.
The Audi Max has a nominal capacity of 800 persons, thus without any kind of organization, the communication and coordination would not work. The traditional approach would be to elect a leader and a group of sub-leaders, etc. in order to build a hierarchically structured organization. Interestingly, the protesters are not organized like this. Instead their movement is self-organized, but structured into several task forces. This makes a lot of sense, since the people powering the protest are not always there - some might lose interest or stay away for some time, while others might join. Thus, while a service, like for example the press information is stable, its establishing components might be not. This is a nice analogy to bodily organs consisting of cells that die and reproduce in a much faster pace than the time span for which the organ is operable.
Furthermore, the students organize two plenum conferences everyday where the task forces coordinate. This plenum is as well based on a grassroots democracy. While there are elected moderators for the discussion, there is no permanent hierarchy in the organization. Yet, the system is effective and stable.
By the way, this is only one out of probably thousands and thousands examples for self-organizing systems. In a paper at IWSOS'09, we investigated how we can design or model such systems.