for internal operations, including troubleshooting, data analysis, testing, research and service improvement,”
via Facebook Manipulated 689,003 Users' Emotions For Science.
“for internal operations, including troubleshooting, data analysis, testing, research and service improvement,”
This testing place in Jan 2012 for – they say – only 1 week ostensibly to gauge reactions to positive or negative things in news feeds. To what ends? To modify newsfeeds to resonate with the user? Isn’t that accomplished by what people and pages you choose to follow? If you follow a page, isn’t that its own indication that you want to see certain posts?
How about letting people make the choices and giving the tools to do so. I think I’m fairly capable of making up my own mind, besides what may resonate with me today, may be different tomorrow. My interests vary often in response to what is or is not going on in my life at that moment. I suspect that’s the case for the vast majority.
Just because someone ‘likes’ a post of woe and pain doesn’t mean it ‘resonates’. It might mean this person is someone I care about and I empathize with what they’re going through – not that I want to be flooded with more related posts of it, or even flooded with suggestions of related posts. Suggesting some is great but I don’t need, nor care for, my feed being taken over. Let me make the decisions.
“for internal operations, including troubleshooting, data analysis, testing, research and service improvement,” — Personally I don’t think this is enough ‘agreement’ to manipulate emotions in the name of testing. Ethics, good ethics, and sensitivity would state up front agreement-not buried in thousands of pages of policy agreement Facebook knows very well most don’t read even if they then highlite it in red. Secondly, inform the person the parameters in the experiment, what they will be seeing and what the purpose of the experiment is. Thirdly, allow active feedback.
But again, Facebook really has no need to even do such an experiment if they understand people can make their own decisions. I do have to say, they have recently asked me on mine whether I want to assist in a study of newsfeeds. The questions were too narrow and poorly constructed, however, and really would not have improved the ‘experience’ for most of us. They were too black and white, too either/or.
Which brings back the question: To what ends was this experiment even needed? And for whose needs?