Facebook users have had their loyalty tested in unusual fashion.
The social network has been deliberately crashing one of its own apps for several weeks to find out whether users would visit its mobile site instead, according to technology business news site The Information.
The app in question was the Facebook Android app and it was reported that users continued to use Facebook via the site rather than give up.
The suggestion is that the unorthodox test is part of a contingency plan to ensure people can use Facebook on Android devices without going through Google’s app store, Google Play, should Facebook and Google, which developed the Android mobile operating system, have a disagreement about how apps function on Android.
Facebook has not commented on the claims, but if they are accurate it would not be the first time it has attempted to monitor the response of users without telling them.
In June 2014, 600,000 Facebook users were subjected to psychological testing without their knowledge or having been asked for consent beforehand.
The social network altered the tone of news feeds by highlighting either positive or negative posts from friends in order to monitor the response of users.
Facebook was able to conduct the experiment because users agree to “internal operations, including troubleshooting, data analysis, testing, research and service improvement” as part of its terms and conditions.
At the time a Facebook spokesman said: “We do research to improve our services and to make the content people see on Facebook as relevant and engaging as possible.”
Facebook recently hit the headlines by creating a feature to warn users if it suspects they are being spied on by government agencies.
The feature, thought to be aimed at the National Security Agency (NSA) in the USA and Government Communications Headquarters (GCHQ) in the UK, is controversial because it does not warn users about other suspicious activity such as hacking.
Facebook has not specified how it identifies the threats either; Facebook’ Chief Security Officer Alex Stamos said: “To protect the integrity of our methods and processes, we often won’t be able to explain how we attribute certain attacks to suspected attackers.
“We plan to use this warning only in situations where the evidence strongly supports our conclusion.”