corona-warn-app / cwa-wishlist

Central repository to collect community feature requests and improvements. The CWA development ends on May 31, 2023. You still can warn other users until April 30, 2023. More information:
https://coronawarn.app/en/faq/#ramp_down
Apache License 2.0
105 stars 14 forks source link

Expand the risk evaluation algorithm so that it also evaluates the number of simultaneous contacts #246

Open Jo-Achim opened 3 years ago

Jo-Achim commented 3 years ago

German: see below.

Cross references to https://github.com/corona-warn-app/cwa-wishlist/issues/213#issue-724544226 and https://github.com/corona-warn-app/cwa-wishlist/issues/240#issue-731384972.

Hello,

with reference to this article “The app needs new features” (SZ from 10/30/2020; https://www.sueddeutsche.de/digital/corona-warn-app-contact-tracing-update-1.5098720) I would like to suggest - if not already done - to expand the risk evaluation algorithm so that it also evaluates the number of simultaneous contacts.

The reason for this is the above Article with a statement made in it (which I do not want to rate): Referring to CWA: "... it was unrealistic right from the start that the measurement method via Bluetooth LE would deliver reliable results." And further: "Since its construction {CWA-App} in spring there have been some scientific findings that could also help the app. Back then, the role of aerosols and super spreaders was not so much in focus. Today it is clear that the risk of infection is significantly higher in environments with many people. "

Assuming the correctness of the statements, I would assume that with the above Extension of the risk evaluation algorithm to be able to map and evaluate the higher "risk of infection in environments with many people".

This approach should be different and simpler than the cross-references above (https://github.com/corona-warn-app/cwa-wishlist/issues/213#issue-724544226 / https://github.com/corona-warn-app/cwa-wishlist/issues/240#issue-731384972).

With this in mind, best regards Jo-Achim.

PS: To avoid misunderstandings: The distance measurements using the Bluetooth signal level should hardly be influenced by software - due to external dependencies such as Bluetooth versions, implementation, transmission power and reception sensitivity, antenna, shielding, etc. Thus, the CWA has to make do with the given circumstances. What should be realizable, however, is an evaluation of the number of simultaneous contacts, so that the findings 'higher risk of infection in larger crowds' can be taken into account; and a kind of disadvantage compensation for Bt distance measurement can be implemented.


Risiko-Auswertungsalgorithmus dahingehend erweitern, dass dieser auch die Anzahl gleichzeitiger Kontakte bewertet.

Querverweis auf https://github.com/corona-warn-app/cwa-wishlist/issues/213#issue-724544226 and https://github.com/corona-warn-app/cwa-wishlist/issues/240#issue-731384972.

Hallo zusammen,

Bezug nehmend auf diesen Artikel „Neue Features braucht die App“ (SZ vom 30.10.2020; https://www.sueddeutsche.de/digital/corona-warn-app-contact-tracing-update-1.5098720) möchte ich anregen – sofern nicht schon geschehen – den Risiko-Auswertungsalgorithmus dahingehend zu erweitern, dass dieser auch die Anzahl gleichzeitiger Kontakte bewertet.

Grund hierfür ist o.g. Artikel mit darin getroffenen Aussagen (die ich nicht bewerten will):
Bzgl. CWA: „…war von Anfang an unrealistisch, dass das Messverfahren per Bluetooth LE belastbare Ergebnisse liefern würde.“ Und weiter: „Seit deren Bau {CWA-App} im Frühjahr habe es einige wissenschaftlichen Erkenntnisse gegeben, die auch der App helfen könnten. So sei damals die Rolle von Aerosolen und Superspreadern nicht so im Fokus gewesen. Heute sei klar, dass die Ansteckungsgefahr in Umgebungen mit vielen Menschen deutlich höher ist.“

Ich würde, die Richtigkeit der Aussagen unterstellt, davon ausgehen, mit o.g. Erweiterung des Risiko-Auswertungsalgorithmus die höhere „Ansteckungsgefahr in Umgebungen mit vielen Menschen“ abbilden und bewerten zu können.

Dieser Ansatz sollte sich von den oben genannten Querverweisen (https://github.com/corona-warn-app/cwa-wishlist/issues/213#issue-724544226 / https://github.com/corona-warn-app/cwa-wishlist/issues/240#issue-731384972) unterscheiden und einfacher sein.

In diesem Sinne mit freundlichen Grüßen Jo-Achim.

PS: Um Missverständnissen vorzubeugen: Die Distanzmessungen mittels Bluetooth-Signallevel dürfte - aufgrund von äußeren Abhängigkeiten wie Bluetooth-Versionen, Implementierung, Sendeleistung und Empfangsempfindlichkeiten, Antenne, Abschirmungen usw. - per Software kaum beeinflusst werden können. Somit muss die CWA mit den vorgegebenen Umständen vorlieb nehmen. Was aber realisierbar sein sollte, ist eine Auswertung der Anzahl gleichzeitiger Kontakte, so dass hier den Erkenntnissen ‚höhere Ansteckungsgefahr in größeren Menschenansammlungen’ Rechnung getragen werden kann; und eine Art Nachteilsausgleich zur Bt-Distanzmessung umgesetzt werden kann.


Internal Tracking ID: EXPOSUREAPP-3549

heinezen commented 3 years ago

Hey @Jo-Achim

Thank you for the suggestion. I have created an internal Jira ticket for your proposal (EXPOSUREAPP-3549). The developers and the RKI will discuss the request as part of the ongoing discussion for cluster detection. We will notify you here on Github if we get any updates regarding the status of this request.

Regards, CH


Corona-Warn-App Open Source Team