3 ms·
>"The CAP and Harvard plans both foresee a digital pandemic surveillance state in which virtually every American downloads an app to their phone that geotracks
by Barrin92 7y ago
>"The CAP and Harvard plans both foresee a digital pandemic surveillance state in which virtually every American downloads an app to their phone that geotracks their movements, so if they come into contact with anyone who later is found to have Covid-19, they can be alerted and a period of social quarantine can begin."
If this refers to the bluetooth supported apps I've seen developed by several institutions, does this not sound like a horrible idea? Let's say two people walk past each other on the street or stand close in a shop, as far as we know the risk of transmission is non-existent or extremely low. Are you just going to build a connected graph and then ping everyone if one person gets sick? You'd be pinging half the population of a dense city after one day.
Before putting these surveillance measures into place I'd like to know if they're not causing more harm and paranaoia then they help.
- SpicyLemonZest 7y agoI don't mean to rip on you, it's important to notice problems, but that seems very solvable. Just require some minimum duration of proximity before you alert.
- Barrin92 7y agoIt's not that easy. Social networks, particular urban dense ones have the characteristics of small world networks (https://en.wikipedia.org/wiki/Small-world_network https://en.wikipedia.org/wiki/Small-world_network). The distance between every person in a city is maybe 5-6 individuals. You can imagine how many hundred people you stand next to, how many people they stand next to, and so on. So if say, the chance of being infected in an interaction is say 5%, then the chance at a distance of two is 0.25%, the chance for someone at a distance of 3 is merely 0.01%, etc.. Do you really think it's a rational use of resources or your attention that if you're one of the countless 'distance 3' people to alert you? Unless you had some mysterious ability to determine with extremely high fidelity that infection has occured from metadata alone, which doesn't exist, it's a tool for mass panic. Bluetooth will give you a match for up to 10 meters. The official safe distance is 6 feet. You can't even tell from a bluetooth match if people were close enough at all.
- SpicyLemonZest 7y agoNo, that doesn't seem rational. Only the distance 1 people should be alerted - that's how it works in existing contact tracing systems.
- mnm1 7y ago> Let's say two people walk past each other on the street or stand close in a shop, as far as we know the risk of transmission is non-existent or extremely low. WTF? What we know is exactly the opposite.
- Barrin92 7y agoWe don't. The primary transmission path is droplets through people in close contact (less than two meters, inhaling aerosols etc..). There is no current evidence that people simply walking past each other can somehow infect others at a meaningful rate, and surface tranmission is negligible as well. See most recent results from the German researchers in Heinsberg.
- mnm1 7y agoWalking past each other and people standing close to each other both imply close contact at less than two meters. Coughing can send the virus around seven meters. You do the math.
- asdff 7y agoAre CAP and Harvard so far in their bubble to realize that not every American has an updated smartphone that can even install new applications, let alone a cell phone at all? This also relies on people having the responsibility to keep their phone charged on their person with the app running at all times.