[chbot] CovidCard implementation - how expensive should it be to manufacture 1M of them?

[Andrew Errington]

If it has bluetooth it doesn't need USB. Set up a HID keyboard profile
and the device could 'type' its data into a text box on a web form.

Use a public/private key pair so that the typed data is encrypted.
Doesn't matter if it's visible during transit, it will only be
decrypted at the destination.

If $1 USB Bluetooth adapters exist then I suggest the cost should be
about $1, especially since the USB parts might not be necessary. The
software itself could run on the Bluetooth microcontroller- the common
controllers from Cambridge Silicon Radio (CSR) are reprogrammable,
although their toolchain is hard to get. On a $1 dongle the controller
might be mask programmed and unalterable.

If you need a dedicated microcontroller you could use the Padauk
PMS150C. It costs three cents (US). I heard they had a more powerful
chip for 15 cents. Maybe that's what we'd need.

I'd run it off a pair of AAA batteries- cheap and ubiquitous.

1) I'd pitch the cost at $2 US
2) I'd allow a month to develop hardware, firmware, website (no app, natch).

However, since the lockdown-by-region is not very granular, I'd just
look at patients' phones and see which towers they'd been close to,
and lock down that area. The CovidCard is probably a smokescreen so
that nobody has to admit quite how accurate cellphone triangulation
is...

Andrew

On 29/08/2020, Bevin Brett <bevin_brett at hotmail.com> wrote:
> Assuming that it is
>
>   1.  Bluetooth is its only radio
>   2.  One USB connection is its only other connection
>   3.  The s/w has only to
>      *   listen continually for nearby devices and remember just their
> presence - no geolocation needed
>      *   broadcast once per minute its own ID and nothing else
>      *   keep track of one month's worth of ID's
>      *   supply its own ID when requested by the USB connection
>      *   supply the recorded ID's when requested by some password-secured
> request over the USB connection
>
> 1 - what do you think the cost per device should be?
> 2 - approximately how long should it take to develop open-source h/w and s/w
> for it?
>
> Given the availability of $1 USB Bluetooth adaptors, it seems to me the main
> problem is a little s/w and an adequate battery...
>