Eben Moglen on the dangers of centralized social networking, in his excellent keynote at The Free and Open source Software Developers’ European Meeting (FOSDEM) 2011 in Brussels, Belgium on February 5, 2011 (full transcript here):
Social networking—that is, the ability to use free form methods of communication from many to many, now, in an instantaneous fashion—changes the balance of power in society away from highly organized vehicles of state control towards people in their own lives.
What has happened in Iran, in Egypt, in Tunisia—and what will happen in other societies over the next few years—demonstrates the enormous political and social importance of social networking. But everything we know about technology tells us that the current forms of social network communication, despite their enormous current value for politics, are also intensely dangerous to use.
They are too centralized, they are too vulnerable to state retaliation and control. The design of their technology, like the design of almost all unfree software technology, is motivated more by business interests seeking profit than by technological interests seeking freedom.
As a result of which, we are watching political movements of enormous value, capable of transforming the lives of hundreds of millions of people, resting on a fragile basis, like for example, the courage of Mr. Zuckerberg, or the willingness of Google to resist the state, where the state is a powerful business partner and a party Google cannot afford frequently to insult.
We are living in a world in which real-time information crucial to people in the street seeking to build their freedom depends on a commercial micro-blogging service in northern California, which must turn a profit in order to justify its existence to the people who design its technology, and which we know is capable of deciding, overnight, all by itself, to donate the entire history of everything everybody said through it to the Library of Congress. Which means, I suppose, that in some other place, they could make a different style of donation.
This problem is directly related to the issue of dominant market share and the high cost of switching services. For an alternative to make any significant impact (like the federated networks that Moglen and others see as crucial), it must integrate with existing services and maintain compatibility, allowing the services to co-exist and the transition to be seamless (more on that in a later post).