[clug] Apple-Google Contact Tracing API. V1 released May 20, 2020

jhock at iinet.net.au jhock at iinet.net.au
Tue May 26 00:41:59 UTC 2020

Please see below:

On 25 May 2020 1:16:46 pm AEST, Hugh Fisher via linux <linux at lists.samba.org> wrote:
>On Mon, May 25, 2020 at 8:16 AM steve jenkin via linux
><linux at lists.samba.org> wrote:
>> The Apple-Google notification service is designed to be anonymous &
>tracking-resistant, without a central database of user details or
>central processing of contacts.
>> There is a central database of rapidly (15-30 min) keys generated on

I'm confused. In one sentence it is explained "anonymous & tracking-resistant, without a central database of user details or
central processing of contacts" and then in the very next sentence, "There is a central database of rapidly (15-30 min) keys generated on devices.". 

 Should this be "There is a database of rapidly (15-30 min) keys generated on devices." or is there a central database that stores the "keys generated on devices."? If it is the former then it seems to be considerate of peoples' privacy. If the latter then people's information is stored on a central database. 


>> App users who are diagnosed Covid-19 positive, notify the App, which
>uploads the last 14 days of keys ’seen’ by the device. The devices with
>those keys are sent notifications.
>> It’s unclear to me how the keys-device connection is made within the
>server database.
>> The Notification side of the server must push a notification to
>devices (presumably when they connect to upload keys).

It would seem that a covid-19 positive person should voluntarily use the app (in our case, written or approved by the Department of Health (DoH) ) to upload the keys, that were on the covid-19 positive person's device, to a central database managed by DoH. The DoH database would then use those keys to notify the devices that contained those keys. 

There must be some way for the DoH central database to map the keys to a mobile phone number which was probably stored on the DoH central database during registration. Doesn't that mean that the keys have to be regularly (possibly "15-30 min") loaded onto the central database for the match to occur and therefore DoH to know whom to notify of the possible infection?

Or, is there a broadband notification of the possibly infected keys to all mobile devices that have the DoH app. Each device will then look for those keys in its own database and if detected then the app would warn the user of possible infection? 

If the former then doesn't that contradict "without a central database of user details or central processing of contacts" because the keys are mapped to a mobile phone number, et al, in the central database?

If the latter then there is possibly more privacy but greater pressure on the mobile networks because of the broadband notifications? Maybe someone has had his or her phone switched off and by the time the notifications arrive those keys have been flushed out of the device's database? 

I can't see how this is any better than the COVIDSafe app except for:

> the [Bluetooth] proximity detection
>is built
>into the kernel as a background task.

My confusion obviously demonstrated. 


>        cheers,
>        Hugh Fisher
>linux mailing list
>linux at lists.samba.org

More information about the linux mailing list