Mubarak's first major attempt to squash the revolution was turning off Egypt's internet on Jan 28. This was shortly followed by an escalation of violence, and Cairo turning into a battleground between the revolutionaries and pro-regime thugs. What happened when the internet went down, and the future of Egypt hung in the balance?
The answer seems to be that the shutoff certainly had a large impact, but it wasn't nearly enough to disrupt the overall network.
By looking at the Twitter timelines of the people in the network from my post Visualizing the New Arab Mind, we can determine which people were affected by the internet shutdown. If the person is tweeting less than half as much during the shutdown in comparison to their normal rate, we color them red. (Some nodes have been removed due to a lack of data)
We can see that the shutdown affected a huge swath of the network, yet those network regions remain dense with active nodes. No one is too far socially removed from someone who still has access to the internet.
We can also take a look at the specifics of when people were tweeting. In the below diagram, we plot every single tweet from this network from Jan 24 to Feb 3. Each node in the network now corresponds to a row of tweets, placed in time. We can see a dramatic cliff on the 28th as service is suddenly interrupted. But soon, many of the nodes are finding a way to tweet at least a little. Cutting the flow of information out of Egypt just wasn't going to happen.
Another notable pattern is the substantial surge in tweets from the people in blue during the blackout period. Clearly a large number of people were alarmed by these developments and increased their tweeting in response. Looking at what they are saying, and if/how they are relaying information from their disenfranchised friends, would be the next step in understanding how the network responded to this challenge.
It's also worth noting that not only were some of the nodes shut off in terms of internet, but also as citizen participants. There were a large number of arrests and temporary detentions, which caused people to go offline. This is also a phenomena worth looking into.