I got the link to work this time, so I've read the discussion. I'm surprised that guy sent you a link to a study that clearly shows homeopathic preparations have no effect. I thought he was trying to convince you it wasn't complete bunk?
I don't think we need to seriously consider that this guy is interested in the challenge. He didn't bother to carefully read the protocol you forwarded there, since he is now off talking about treating disease, rather than provings. And FYI, taking the same drug 50 times does serve as independent trials - for example, a new cold medicine could be tested by taking a course of the drug at the start of every cold and recording the duration of symptoms. If each course consists of either active treatment of placebo (randomly assigned and blinded), the times when you took placebo and the times when you took active treatment could be compared in the same way that 50 different people taking the drug at one time could. It's just that it would take more than a decade to take the treatment 50 times and 'time' would be a confounder.
First, let's looks at the study.
http://www.smeddum.net/downloads/2005_Homeopathy_Proving_Symptoms_Nonlocal.pdf
It shows exactly what we'd expect if homeopathy has no specific effect. There were no differences in the amount and kind of symptoms experienced by those on placebo and those on homeopathic preparations. A homeopath was unable to tell what preparation was taken based on the symptom list, even when given a list of twelve possibilities. There were more symptoms experienced whilst taking the homeopathic/placebo preparation than there were in the two weeks before the preparation was given. All they actually managed to prove was that when you tell the homeopath what symptoms to expect, they are more likely to find those symptoms from amongst the mix. Lovely, lovely example of how non-blinding leads to bias.
Now, to a reasonable person, this would be a nail (several actually) in the coffin of homeopathy. However, they've totally spun this around in their discussion to conclude that homeopathy does have an effect. First of all, they assume that the increase in symptoms has to be due to a specific effect. It seems they've never heard of the power of suggestion or the 'nocebo' effect. Then they assume that the effects of homeopathy are non-local. That is, "the remedy and the whole of its preparation history,
together with the homeopathic experience that contains and produces the effect, and not only the remedy as such", so that even the placebo group partakes of this homeopathic experience. How else could you possibly explain the increase in symptoms in the placebo group, and that the placebo group was indistinguishable from the group taking the preparation?
So it looks like Yuri's test has already been done, and the excuses for the abject failure have already been wrapped up in a nice little package with a bow on top. So I propose a new test.
It follows the same procedure as the study. As many volunteers as possible are recruited to take an unknown preparation (chosen randomly from the list of twelve) or placebo. Diaries are kept for 1 week as a baseline (only for the purposes of looking for baseline differences in patient characteristics), and then for 2 weeks after the preparation has been given. The diary entries are then coded as described: "After the proving, the symptoms were saved in an electronic database. Symptoms were coded individually with onset, duration, location, sensation, extension and modality. The symptom database was arranged according to the head-to-foot-scheme, randomly shuffled so that entries were not discernible by volunteer or day..." However, instead of passing these entries on to one materia medica expert, they are passed on to as many as possible. Except half (chosen randomly) are told the correct name of the preparation, and half are told it was a different preparation (deliberately chosen from the list of twelve to have symptoms distinguishable from the correct preparation). It is important that these experts
do not know (or know that it is possible) that they have been given incorrect information. Then the results can be compared for any differences between preparation and placebo, and any differences in the presence of typical and atypical symptoms between experts given the correct name and the incorrect name. If no differences are found, it confirms that the results are simply the result of cognitive bias, rather than anything specific to the homeopathic experience.
The problem is, the trial is no longer as simple, and can't be done by an individual. And it's possible that they would come up with an even more idiotic excuse to explain away their failure, and my head would finally explode.
I can come up with a way for an individual to do this, but realistically, it involves some dedication and the desire to actually test your ideas. And I really don't see that sort of sincerity from anyone in the homeopathy camp, or the woo you are arguing with.
Linda