Jim Lesurf wrote:
[snip]
Arny has reached certain conclusions from a test. That test relied on a
certain method. And that method - whether he or anyone else like it or
not - arose from a particular methodology. In very plain terms I was
asking for the reasoning behind the method.
Is this specific to the individual test(s) he has described? Or are you
asking about the method generally called 'ABX' whenever it is employed? My
impression is that you are directing your questions just to a specific
instance, but I am not entirely sure of that.
It *shouldn't* matter - methodology and method can and in many cases
should change in successive stages of experimentation. But one method
draws on one methodological approach - so in that sense I'm interested
in the methodological basis of ABX.
However whatever his view, it may not alter the actual methods or
results he and others refer to.
I can only assume that he doesn't have a view.
FWIW In my experience many academic scientists and engineers employ the
scientific method and various experimental protocols because they are the
usual techniques they are taught and find useful. Many seem not to concern
themselves with the arguments for or against them. Just use the tools from
the toolbox. I doubt most of my ex-colleagues would know what
'epistimology' or 'ontology' means without looking it up. They would
suspect they have encountered a theologian, or a philosopher who walked
into the wrong dept by mistake. :-)
You're experience is probably quite representative. I was relaying this
episode to a friend - a polymer scientist - and he was with me up until
we started to work back to discuss ontological 'realities'. I don't
pretend to understand this whole issue - more of a working familiarity.
The difference between 'lab science' and the situation we're discussing
is human interaction on a far from basic level (hearing/listening for
example) with physical 'facts', and I think there may be methodological
anomalies.
If you are asking for a more general
explanation of something anyone might give, then perhaps someone else
can help. None of this was/is clear to me, hence my question.
No, nothing general - just why he would choose a method for a test. I
wasn't asking for general answers - it's by belief that there is no
'correct' methodology.
Can you explain what you mean here by "correct"? Your wording implies a
unique methodology.
Correct is something that can be applied consistently. If a theory is a
way of explaining the world's workings (social, political and physical),
the methodology is the basis of theory. The problem - and hence no
'correct - is that people can arrive at the same theory from different
methodological positions.
The reality is that various techniques may be applied,
and are chosen on the basis of what idea(s) an observation or experiment is
aimed at testing, and what forms of problems may be significant in the
specific context.
The context was several 'facts' Arny laid out earlier in this thread.
Again, as in my previous posting re 'context' - I don't know what
'facts' you are referring to here. If your point is specific, can you
please explain?
The specific point, and where this thread started, was an assertion that
CD-standard recording captures the whole LP audio recording for all
practical purposes. I had certain issues with the source of that
assertion which went unanswered; no matter. Arny then associated that
assertion with certain facts:
1 - CD-standard recording captures the entire LP music recording. 2 -
CD-standard recording captures in entirety any variance in sources.
As you inserted yourself, the above need to be qualified in some way to
mean they refer to what is 'audible' in terms of being distinguishable
since all real systems will have limitations. Ditto for the circumstances
of use. However I would take such qualifiers to be read into the statements
in this context.
That said, the above seem simply to re-state the assertion you questioned.
However is this not on the basis that controlled tests return results that
support these "assumptions"? So your point is to question the nature of
those controlled tests?
They do - I wouldn't call them facts by the way - hypotheses. I could
count them as facts if I knew the methodology and method.
When a statement has been subject to controlled experimental tests,
designed to cope with the relevant experimental problems, and found to be
supported, then the conclusions should only be called "assumptions" with
care as this term might me misunderstood. Do you do this because you don't
know the details of the experiments or the results?
The method is, I'd guess, pretty well thought through in many
significant respects. I questioned the sample, but Arny didn't respond.
I could easily say that if I hold a pen and then let it go it is my
"assumption" that it will accellerate downwards and fall to the ground.
However most people in most normal circumstances would not feel that
calling this an "assumption" means it is a mistake or in any serious doubt.
Of course, I can find circumstances where it won't apply, and in general,
we can expect any conclusions to only apply within a range of
circumstances, etc.
No, I'd let that go so to speak :-)
In general, also, if you have doubts about a given experimental design,
etc, and regard the results as doubtful, the normal recourse in science is
to propose better controlled experiments and judge on the basis of their
results.
I felt these were assumptions,
I obviously can't speak for Arny, but my understanding is that suitable
tests do support what you call assumptions.
I'm sure they do - even I wouldn't count my experience (to the contrary)
as valid data. It just got me thinking, that's all.
Also that descriptions of the
experimental designs and the control conditions, etc, have been dicussed on
many occasions over the years. Given this, is it suprising if Arny decides
he can't be bothered to cover old ground yet again? Is this not already
covered on his website or elsewhere?
It's not the conditions etc relating to method that I'm bothered about
until I can establish the methodology. I mentioned elsewhere that I
guess it's positivist or empiricist - but I can't be sure. As I said
earlier, different methodologies can lead to similar methods.
and Arny then led me to a test carried
out which I think he feels was a good example of data collection in this
context:
http://www.provide.net/~djcarlst/abx_digi.htm
[snip]
Digging a little deeper, there's a reference on the ABX site to
something called "Virtual Reality Methodology". I wondered what that
methodology was all about. Arny wouldn't tell me.
Afraid I don't know off-hand what it means, so can't comment on that. :-)
If I have seen the phrase in the past, then I am afraid I have forgotten
about it. But when I get a chance I'll check the above reference.
Arny did say there was a fuller explanation on the site, but I haven't
been able to find it.
Rob