Rossi is famous for his demos, which are often conflated with experiments when people are arguing in a hurry. But there have also been several reports based on actual tests involving somewhat careful protocols; e.g., the Ferrara test. Were they scientifically adequate? I can only conclude from the comments I've followed that either these tests were inadequate to establish their conclusions because of some methodological flaw (e.g., a bad or missing control); or, at minimum, there were one or more holes that were in need of filling in during a subsequent pass.
Is a "test" an "experiment" in all of the ways you mention, oldguy ? Perhaps not. But neither is it a lackluster demo of the kind seen recently in Stockholm. Enough of those more rigorous tests would justify taking interest in something provided there is follow-through. So I do not fault many of the people with relevant credentials who publicly showed interest except for a lack of caution, given all of the negative indicators that have long accompanied Rossi.