Alexander's article was convincing because the downside of ignoring the regulations in his case was pretty close to zero. Giving a questionnaire to people prob

Alexander's article was convincing because the downside of ignoring the regulations in his case was pretty close to zero. Giving a questionnaire to people probably isn't going to hurt them. Injecting viruses into desperate people, on the other hand, has some pretty obvious risks. One can argue whether the current amount of regulation regarding it is excessive, but its pretty clear there needs to be some regulation. Plus, note that the problems here weren't only regarding safety issues. The study was also poorly designed from the standpoint of getting meaningful data out of it. So even if one was indifferent to the safety of the study participants, the whole exercise seems to have been close to worthless, since the data is unusable.