The corona app uses Bluetooth to keep an eye on who you have been in the vicinity and only works if both people have installed the app. Based on Bluetooth signals, the app sees whether, how long and how close you have been to someone else. Use of the app is voluntary.
Thousands of reports
In recent times, an average of 600 positively tested people reported their infection via the app every day. This would mean that every day many thousands to tens of thousands of people should receive a warning from the Corona Detector.
Yet many people are not notified. “This may have to do with the group of people who use the CoronaMelder”, says Ron Roozendaal, director of information policy at the Ministry of Health, Welfare and Sport (VWS). “The CoronaMelder users are often familiar with the measures and are more aware of corona.”
About ten percent of the people who get tested after a report in the CoronaMelder is actually infected. A large proportion already had complaints, a somewhat smaller proportion had no complaints. About 15 percent of all people with complaints turned out to be infected, and about 5 percent of people without complaints.
According to Roozendaal, almost none of them had been approached by the GGD in the context of regular source and contact tracing. “In short, CoronaMelder warns people more quickly.”
The makers of the app are positive, partly because it appears that the dreaded adverse effects of the CoronaMelder app have not been achieved. This concerns, among other things, privacy concerns and compensation behavior: the fear that the app would unjustifiably feel safe and no longer adhere to other measures. These concerns appear to be unfounded, the government states on the basis of both samples and surveys.
Sometimes users get a warning from the app even though they haven’t been around others at all. “We also know those stories”, Roozendaal explains. “That is why we are now, together with Germany, conducting a test with robots and barrels of water that people are supposed to represent, and placing obstacles such as walls and windows between them.” For example, it must be clear whether users of CoronaMelder can receive false reports through windows or walls.
The GGD also does not always ask positively tested people whether they want to pass on their contamination via the CoronaMelder. “It is part of the range of duties of call center employees, but some have not yet followed the training on the CoronaMelder, which means that this is sometimes forgotten”, says Roozendaal.
According to the GGD, all call center employees have now completed training on the app.
Open and transparent
The ICT assessment advisory board, the government’s ICT watchdog, will evaluate the development of the CoronaMelder. The research is mainly about the openness and transparency with which CoronaMelder has been developed. Any developer, designer or other specialist could help with the development.
This way of working is completely different from many government ICT projects, which are often developed behind closed doors by ‘dearly paid external parties’ and are therefore ‘unnecessarily or unnecessarily large’, according to the same watchdog.
At the CoronaMelder, everyone could watch and help from the start. “I am very proud of that”, says Roozendaal. “From the source code to the pen tests (reports with vulnerabilities found by ethical hackers, ed.): Everything is public with us. And we now receive several questions from the government to present our work and development process to them.”