Imo - the most important reinforcement is not whether it accurate within 100 feet vs. 10inches as most of the fact checks focus on. We don’t know what provider they used and there are different methods of collection, some of which are less accurate than your GPS and some more. Here’s a blip from one provider:
“ The code that collects the location data uses all available sources to geolocate the phone: cell tower signal; GPS signal; WiFi; and Bluetooth. Note that the phone does not need to be connected to either WiFi or Bluetooth in order for it to be triangulated using those methods; the only thing necessary is for the WiFi and Bluetooth sensors, respectively,of the phone to be turned “on”. Because of the extended triangulation methodology, location data collected in this manner are several times more accurate than the methods above.”
The argument in the fact checks is that CELL TOWER data is not precise enough - with an error of something like 10 meters. Saying that GPS/Bluetooth IS accurate doesn’t refute that point. Rather I think:
we don’t know which method of tracking was used, most providers use all of them. You can’t assume it was the least accurate type, but even if it was cel tower data, that’s still accurate w/in 30-50 ft.
“We’ll it’s an urban area, they may have just been sort of near the drop box” okay, but this is why they filtered only devices with 8x instances. Being 50ft away from 8 different drop boxes over a large geographic area and visiting an NGO removes almost all risk of just picking up someone who happened to be near a drop box
“we’ll they could be a bus driver” okay maybe it’s 2000 bus drivers, but then there is video matching. I do want to see the data, it should be easy to tell whether the decides were incidentally driving a lot and passing drop boxes vs. stopping and lingering. This should be auditable and supposedly data is coming.
“they didn’t include any video showing the same person at 2 boxes” this is imo the strongest counter point. Since we don’t see the video-matched people visiting multiple drop boxes per person, you CAN make the argument that we need to see the data. This was my biggest gripe with the movie - why didn’t they show the video multiples (we know many vids were deleted and maybe the deletions we’re strategic so that no one visits 2 videod boxes, but that seems like a bit of a stretch). Also why didn’t they name the NGOs or confront any mules?
I think Mules is the most digestible-for-mainstream proof of fraud yet, but not the most robust. It should be the least dismissible - the methodology of a threshold multiple geofences more than offsets the precision within the geofence even if they used the least precise method. The best argument to make is the threshold for multiple visits and the video. The biggest hole is the lack of multiple-mule-vid-captures.
The main commercial use for finance (I don’t know about advertising) is:
Hedge funds: Predicting retail company sales trends by tracking aggregate foot traffic in Best Buys or predicting Tesla unit production by tracking average number of employees in their factories for the quarter
Commercial Real Estate: Tracking general demographic trends and traffic around a building
Advertisers: I don’t really know how they use it, it think to market geographically relevant ads
The rules are evolving and Europe attempted to implement some privacy protections but basically every app just made you check “Opt In” to use then at all. Technically it is “deidentified” and only attached to a device and anonymized marketing ID. But some ad companies match the marketing IDs of a bunch of different data types (location and other) into one unified profile. And if someone cares to find out which marketing goes to your house every night and your workplace every day, well there’s a good chance that marketing ID is you.
I don’t think the microphone/camera in your phone are passive, but I think it’s best to assume everything else searches and accounts and keystrokes and location and habits are all available if you’re even important enough to draw govt persecution.
Kinda makes you wonder how any high profile murder goes unsolved huh? I guess the powers that be don’t want to jeopardize it over “small things” and risk losing the tool where it is most valuable - hopefully that is in foreign/military ops but probably it is blackmailing pols and manipulating elections to protect big tech.
Imo - the most important reinforcement is not whether it accurate within 100 feet vs. 10inches as most of the fact checks focus on. We don’t know what provider they used and there are different methods of collection, some of which are less accurate than your GPS and some more. Here’s a blip from one provider:
“ The code that collects the location data uses all available sources to geolocate the phone: cell tower signal; GPS signal; WiFi; and Bluetooth. Note that the phone does not need to be connected to either WiFi or Bluetooth in order for it to be triangulated using those methods; the only thing necessary is for the WiFi and Bluetooth sensors, respectively,of the phone to be turned “on”. Because of the extended triangulation methodology, location data collected in this manner are several times more accurate than the methods above.”
The argument in the fact checks is that CELL TOWER data is not precise enough - with an error of something like 10 meters. Saying that GPS/Bluetooth IS accurate doesn’t refute that point. Rather I think:
we don’t know which method of tracking was used, most providers use all of them. You can’t assume it was the least accurate type, but even if it was cel tower data, that’s still accurate w/in 30-50 ft.
“We’ll it’s an urban area, they may have just been sort of near the drop box” okay, but this is why they filtered only devices with 8x instances. Being 50ft away from 8 different drop boxes over a large geographic area and visiting an NGO removes almost all risk of just picking up someone who happened to be near a drop box
“we’ll they could be a bus driver” okay maybe it’s 2000 bus drivers, but then there is video matching. I do want to see the data, it should be easy to tell whether the decides were incidentally driving a lot and passing drop boxes vs. stopping and lingering. This should be auditable and supposedly data is coming.
“they didn’t include any video showing the same person at 2 boxes” this is imo the strongest counter point. Since we don’t see the video-matched people visiting multiple drop boxes per person, you CAN make the argument that we need to see the data. This was my biggest gripe with the movie - why didn’t they show the video multiples (we know many vids were deleted and maybe the deletions we’re strategic so that no one visits 2 videod boxes, but that seems like a bit of a stretch). Also why didn’t they name the NGOs or confront any mules?
I think Mules is the most digestible-for-mainstream proof of fraud yet, but not the most robust. It should be the least dismissible - the methodology of a threshold multiple geofences more than offsets the precision within the geofence even if they used the least precise method. The best argument to make is the threshold for multiple visits and the video. The biggest hole is the lack of multiple-mule-vid-captures.
WTF, do they really log all this info? How do I stop them from tracking me? This should he illegal.
The main commercial use for finance (I don’t know about advertising) is:
Hedge funds: Predicting retail company sales trends by tracking aggregate foot traffic in Best Buys or predicting Tesla unit production by tracking average number of employees in their factories for the quarter
Commercial Real Estate: Tracking general demographic trends and traffic around a building
Advertisers: I don’t really know how they use it, it think to market geographically relevant ads
The rules are evolving and Europe attempted to implement some privacy protections but basically every app just made you check “Opt In” to use then at all. Technically it is “deidentified” and only attached to a device and anonymized marketing ID. But some ad companies match the marketing IDs of a bunch of different data types (location and other) into one unified profile. And if someone cares to find out which marketing goes to your house every night and your workplace every day, well there’s a good chance that marketing ID is you.
I don’t think the microphone/camera in your phone are passive, but I think it’s best to assume everything else searches and accounts and keystrokes and location and habits are all available if you’re even important enough to draw govt persecution.
Kinda makes you wonder how any high profile murder goes unsolved huh? I guess the powers that be don’t want to jeopardize it over “small things” and risk losing the tool where it is most valuable - hopefully that is in foreign/military ops but probably it is blackmailing pols and manipulating elections to protect big tech.