[toby]

Not drinking Joe

Why the hell not?

But on that note, if the absolutist says a test is invalid due to the samples not being absolutely the same, how do they know they aren't?  Probably they are not, but its possible they are, until tests are done to prove otherwise. How fair is it to bunk findings based on a probable assumption? Especially when its not known that those microscopic differences would be perceptible.

That's why it all depends on what you're going for.  I take good notes and follow as consistent processes as I can manually, and my results may not be exactly repeatable if you did some chemical analysis.  But they are close enough for my purposes.  After all, most of the tests at places like brulosophy.com are specifically geared towards whether the average person can tell a difference between a couple of variables.  They're ultimately geared towards finding out if there's any need to worry about technique X vs. RDWHAHB.  It's more like the Pepsi challenge than a rigorous scientific experiment.

[klickitat jim]

Precisely.  And the new findings thread by Marshall about DMS is another one that points out this stuff. For me, Marshalls findings are pretty solid evidence, at least enough proof for me to feel safe trying it with my beers on my system. But,,, others might point out that he didn't present foundation for the evidence. The scientist wasn't here, we didn't get to question his qualifications,  or cross examine his testimony, so its all just inadmissible hearsay. Sometimes you cant win for losing, LOL. Poor marshall.

[denny]

Precisely.  And the new findings thread by Marshall about DMS is another one that points out this stuff. For me, Marshalls findings are pretty solid evidence, at least enough proof for me to feel safe trying it with my beers on my system. But,,, others might point out that he didn't present foundation for the evidence. The scientist wasn't here, we didn't get to question his qualifications,  or cross examine his testimony, so its all just inadmissible hearsay. Sometimes you cant win for losing, LOL. Poor marshall.

Jim, always keep in mind...this is homebrewing.  We're not trying to cure cancer here.  For me, that means that the standards for evidence are much lower, and rightfully so.

[Phil_M]

I think that without debate, there isn't really science. A good scientist is going to question everything. Isn't that part of why many scientific jobs require degrees? School is where you debate about which direction a friction force is applied in or how different parts of a mechanism move in relation to each other.

I'm not saying that some debates aren't easy to solve, (See the friction reference) but the point is the debate still exists.

Now we can debate on if that statement makes any sense or not...

[klickitat jim]

Precisely.  And the new findings thread by Marshall about DMS is another one that points out this stuff. For me, Marshalls findings are pretty solid evidence, at least enough proof for me to feel safe trying it with my beers on my system. But,,, others might point out that he didn't present foundation for the evidence. The scientist wasn't here, we didn't get to question his qualifications,  or cross examine his testimony, so its all just inadmissible hearsay. Sometimes you cant win for losing, LOL. Poor marshall.

Jim, always keep in mind...this is homebrewing.  We're not trying to cure cancer here.  For me, that means that the standards for evidence are much lower, and rightfully so.

For me too

[erockrph]

My biggest takeaway from all of Marshall's exBeeriments is that it's well within the ability of a good homebrewer to produce consistent/repeatable beer, since so many triangle tests fall outside the level of statistical significance. The exbeerimenters are able to produce beers within a tolerance such that even when making a small change the beers are virtually indistinguishable.

To echo the sentiments of others, "same" has different tolerances depending on what you're testing and the resolution of your measuring device. For a beer tasting, the ideal measuring device is my own palate. Any other measurement has to be viewed with the understanding that I may not get the same results. At that point you need to decide how much weight to give those results based on how much you trust the source and whether it falls in line with your own experience. In those cases, I am unable to assess the "sameness" of the samples directly, so the more/better data the higher my confidence is in the results.

[leejoreilly]

To echo the sentiments of others, "same" has different tolerances depending on what you're testing and the resolution of your measuring device. For a beer tasting, the ideal measuring device is my own palate.

"You don't need to measure with a micrometer if you're going to cut with a chainsaw."

[hopfenundmalz]

To echo the sentiments of others, "same" has different tolerances depending on what you're testing and the resolution of your measuring device. For a beer tasting, the ideal measuring device is my own palate.

"You don't need to measure with a micrometer if you're going to cut with a chainsaw."

In the engineering lab's there was a saying that went like this.
Measure with a micrometer, mark it with chalk, cut with an axe, paint it to match.