I was actually hoping to learn more about how blind tests, single, double, or whatever the flavour, is conducted and how and why their's is the only true method. Further, it would be nice to know more about what it is that is measured scientifically to prove the superiority one component over another similar component, especially cables (since many audiophiles seem to bear a grudge on obscenely priced cables smeared with snakeskin oil).
I read Stereophile on a fairly regular basis, and believe that their mix of subjective reviews accompanied by measurements (in many reviews, though not all reviews), is a sound approach indeed. By the way, across the pond on the other side of the Atlantic, HiFi News and Record Review does much the same thing, though only a very small subset of their measurements are actually published in the print version, the remaining being made available on their website.
I don't understand the sudden reticence when asked to come up with scientific and objective methods and data. Nevermind, the thread is still open in anticipation of outpourings of information and knowledge that will hopefully educate us all.
PS: J Gordon Holt who has been quoted here is a well respected and successful editor, but his words are not considered the only Gospel anymore, if one is to go by recently published (in Streophile) opinion of one of its most characterful writers (Dudley, who else).
PPS: I still think a judicious mix of subjective and objective is the way to go (though I'm jumping the gun here).
There is a very easy way to experience the blind test.
Download foobar 2000 audio player.
foobar2000
Then download the ABX plugin into the plugins folder:
foobar2000: Components Repository - ABX Comparator
Right click two tracks (say one is wave file, and another one is mp3), and take the test.
ABXY test goes like this:
A and B are reference materials.
X and Y are unknown materials.
You have to tell whether X is A, B is Y
OR
X is B, Y is A
ABX test - Wikipedia, the free encyclopedia
So how would one go about doing the same for Snake oil?
One circuit will have snake oil, another circuit path will not have snake oil.
Thus you establish two references.
And then you have another circuit "X" which may have snake oil / or may not (this will change randomly with trials).
You have to identify whether the circuit X is snake oil one (by comparing snake oil reference) or non-snake oil (by comparing to non-snake oil reference).
Its not that complex - all you require is wires and switches.
So what exactly is measure SCIENTIFICALLY?
Its the confidence levels.
If you get 10 rights and 10 wrongs - you are guessing.
If you get 20 rights and 0 wrongs - you are not guessing and there is definitely a difference.
What happens if you get 0 rights and 20 wrongs?
You could hear the difference and did it wrong on purpose.
***
If you wish to do a DOUBLE blind test, then you should not even know what the reference A and B are. That is you shouldn't even know that A = snake oil, B = non snake oil. You will know them just as references. And then go about matching X with A or B