Sorry, but neither point is convincing to me. First, I don't get why sitting in one's bedroom and writing in a public forum illustrate the clear distinction be

Sorry, but neither point is convincing to me. First, I don't get why sitting in one's bedroom and writing in a public forum illustrate the clear distinction between public and private spaces. You already noted the forum is public. IT IS PUBLIC. If you put up your web page on the Internet and only search engine would ever come to your page, well, your page is still the public domain. Being in a small community, take a cult, well, if you show up in the public and doing your activity in the public, I would expect that to be a public demonstration. In fact, if you try to run naked with your cult in the middle of the street, you are going to get arrested and charged for public misconduct. If you are running naked in a cult house, well, that's private. The distinction is very clear in these examples. I am not arguing there aren't examples that would illustrate a blur, but your example is too weak. My response to the second suggestion. Well, leak of data is a big deal. So if my research somehow harms people with AIDS, am I allowed to collect the non-AID patients? Another way to put it, filtering doesn't do any good and provides no meaning. From an ethic standpoint, people aren't concerned about what groups will be harmed from the outcome of the research - because the whole point of research is to either validate or invalidate some hypothesis. Research ethic should be about the treatment of data and the process of conducting experiment. However, I do agree that some users aren't aware of what online service do with their application.