You're comparing the panicked, reactive actions of officials in the chaos of Libya, Egypt and Tunisia to the stable, proactive, organized security apparatus of the United States? Really?
Let me try to address your points one at a time.
1) In Boston, they shut down the entire city because they didn't know where any more devices would have been placed or what fake name they may have been registered under - not because they didn't have any more granularity. If you're a dictator looking to prevent word of a massacre from getting out, there are very good technical reasons for blocking entire sites at a border-router level: least-work principle, speed, and thoroughness to name a few. Even the loyalty of senior officials was in question (considering how many of them defected or fled) - why not just shut down social media for everyone rather than try to figure out who's loyal enough to be trusted with it?
2) Such technology already exists. If you have access to a network operator's systems, you do not need to know the phone number in order to contact a phone. A phone number, like a domain name, is a convenience of addressing for the benefit of humans that glosses over the underlying network structure. Ever activated a prepaid phone by calling the activation line or going to a website, reading them the IMEI under the battery, then waited for the phone to come to life? You would have witnessed the reception of activation and control data when the device didn't technically yet even have a phone number.
3) Hell, even inspecting the frequency bands and/or encoding schemes in use can do a reasonably good job of determining the service provider used by a particular phone. The location (to within a few feet) and technical details of every cell-phone antenna in the country are registered with the FCC. It wouldn't take much to do a quick query that locates all the mobile phone antennas within receiving range of a demonstration, use some agency's cozy relationship with the telecomms companies to query the computers in charge of handling connections to those antennas (thus obtaining an instant list of most people at the demonstration), and send the kill commands to every phone on that list that isn't on some kind of other whitelist. Do you believe Verizon, AT&T, T-Mobile and Sprint have no idea who's connected to their towers (and thus couldn't bill you for roaming)? Even if you're using a virtual network operator, the physical network operator (of which there are not many) must, by definition, know enough about who is connected in order to establish a control connection to your device to route calls.
4) Or, you could just do things the short way. Take your list of nearby cell antennas and physical network operators and use your back door into each to send a broadcast command to everyone on the entire tower without ever having to know or care who was connected - the "shoot them both and sort it out later" approach. Cell broadcasts like this are already done for warning systems (e.g. earthquake detection in Japan) and reverse-911 calls.
= new reply since forum marked as read
