Moderation and placebos

Bus No. 1985: You need less shit.

When is a placebo a placebo, and when is it not?

Pour example, placebo buttons. There's an intersection near my house: There's a pedestrian crossing button there, which sometimes 'works' and sometimes doesn't. By works, I mean that if you push it, it stops traffic and changes the little red man to a little green man when the time to cross rolls around. Depending on the time of day and levels of traffic, however, it doesn't actually change the traffic light timing; it merely changes the little red man to a little green man when the time rolls around. That first clause is important; depending on the time of day and level of traffic. If you push the button late at night or early in the morning or really, any other period of time when traffic is light, then it actively changes the traffic light and changes the little red man to a little green man within seconds. Push the same button when traffic is heavy, and prepare to wait. So, is this a placebo button or not? Sometimes it is, and sometimes it isn't.

That's example one. Second example is about pharmacological placebos. I'm thinking of a case whereby it's possible to prime or boost the effectiveness of placebos through pseudo-homeopathic means Imagine a case whereby patients are divided into three groups to trial out a new drug. One group is given a traditional placebo, one group is given the drug, and one group is given something in between; they are given the drug, but not enough of it i.e. the dosages they are given are too small to be effective. I'm wondering what the results would look like, and whether this is too spookily homeopathic to try.

Gettier placebos? Functionalist placebos? What else could you do?