While these active MitM attacks are important (the methods seem to be similar to ARP-poisioning), we shouldn't leave out the passive capabilities. These may not even be listed as a feature, if it is a different tool that parses the already-captured traffic as a deferred job.
As we see mentioned here on HN all the time, there is a massive amount interesting data that can be pulled out of large datasets. The original WP publication[1] about COTRAVELER gives a very nice example of the power in just knowing very-inaccurate (cell-sized) location data. You probably don't even need any particular cell-network identifying number, given how easy it is to correlate this kind of data to other identifiers.
Comments
While these active MitM attacks are important (the methods seem to be similar to ARP-poisioning), we shouldn't leave out the passive capabilities. These may not even be listed as a feature, if it is a different tool that parses the already-captured traffic as a deferred job.
As we see mentioned here on HN all the time, there is a massive amount interesting data that can be pulled out of large datasets. The original WP publication[1] about COTRAVELER gives a very nice example of the power in just knowing very-inaccurate (cell-sized) location data. You probably don't even need any particular cell-network identifying number, given how easy it is to correlate this kind of data to other identifiers.
[1] http://apps.washingtonpost.com/g/page/world/how-the-nsa-is-t...