From: grahn+nntp@snipabacken.se   
      
   On Sun, 2020-04-26, skybuck2000@hotmail.com wrote:   
   > On Saturday, April 25, 2020 at 6:41:36 PM UTC+2, Jorgen Grahn wrote:   
   >> On Sat, 2020-04-25, skybuck2000@hotmail.com wrote:   
   >> > Corona Update 4 (25 april(4) 2020):   
   >> ...   
   >>   
   >> > 8. The corona app idea to inform people of infections around them is   
   >> > no success so far, to much security/privacy concerns. I already   
   >> > thought about such apps long time ago and discussed it with my   
   >> > mother. She found it a bad idea, and me too actually because once   
   >> > you go this route you can start creating all kinds of   
   >> > "snitching"-apps and then our privacy goes out of the window.   
   >>   
   >> The Pirate party showed[0] a solution which would respect privacy.   
   >> You have a phone app which roughly:   
   >>   
   >> 1. Sends out a series of random numbers over a short-range   
   >>    communication medium.  (I don't own a smartphone so I don't know if   
   >>    this already exists.)   
   >>   
   >> 2. Receives and stores others' random numbers.   
   >>   
   >> 3. Receives messages like "the guy who sent  random numbers has   
   >>    been diagnosed with a transmittable disease" and notifies the user   
   >>    if the app has seen a lot of them.   
   >>   
   >> No central database, and nothing meaningful to sniff for in public   
   >> places (not until you get ill anyway).   
   >   
   > Seems the idea here is to avoid "identification numbers".   
   >   
   > Seems like a single person has multiple random identification numbers.   
   >   
   > There random indentification numbers would need to be very big   
   > others there is too much potential for conflicting and confusing   
   > random numbers.   
   >   
   > Also the messages in which these are "broadcast" I presume would   
   > have to be "anonymous". So no source ip and such.   
      
   Yes.  The numbers themselves carry no information, but they would have   
   to be anonymous.  Once again, I don't know if this mechanism exists.   
      
   > There is one problem with it though.   
   >   
   > Let's suppose random numbers X,Y,Z are from an infected person.   
   >   
   > How does infected person re-broadcast these numbers to inform that   
   > this person is infected.   
   >   
   > There needs to be some information stored about source/dest/contacts   
   > to be able to do that.   
      
   I think my (3) above would be "download today's list of infected   
   numbers from the national health authority".  If none or few of them   
   match numbers received in (2) you can discard the information.   
      
   But I didn't read and understand all the details.  The main point was   
   that it's possible to implement a virus alert mechanism that doesn't   
   violate privacy.  Provided (1) exists in phones, or in some other   
   equipment people are willing to carry.   
      
   /Jorgen   
      
   --   
     // Jorgen Grahn    O  o   .   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)