Proposal to improve the ROBERT protocol

[bobiciel]

Hello,

In order to kill the containment time and because the intellectual exercise seemed interesting to us, we tried to develop a solution that would improve the ROBERT protocol.
http://bobiciel.com/cariboost_files/Proposition_d_un_protocole_de_suivi_des_contacts_rapproches_dans_la_lutte_contre_le_COVID-19.pdf
(sorry, only in French at the moment)

The specifics of our proposal are as follows:

• Pseudonyms (single-use identifiers) are generated by the application installed on the client device.
• Only potentially contaminating single-use identifiers are stored on the central platform.
• Offuscation is added when the central platform is queried by a client device.

Thank you for your feedback

Sincerely

Bob

thanks Deepl for the translation ;-)

[vincent-grenoble]

Hello @bobiciel. Thank you for your feedback, I gave a look at the document.
The following item:

2. Lorsque Alice présente des symptômes du Covid-19 :

• Alice signale sa possible contamination à la plateforme centrale.

reminds me of the latest article of Oxford epidemiologists [ref1] (a MUST read) that I mention in Issue #59. The authors suggest having a quick information, even prior to having a positive test result. Something that could be discussed, but that also requires a careful procedure to avoid database pollution.

You would probably also be interested in having a look at our new DESIRE proposal for a interoperable 3rd-way exposure notification:
https://github.com/3rd-ways-for-EU-exposure-notification/project-DESIRE

Cheers.

ref1: "Digital contact tracing: comparing the capabilities of centralised and decentralised data architectures to effectively suppress the COVID-19 epidemic whilst maximising freedom of movement and maintaining privacy."