Archived posting to the Leica Users Group, 2000/01/23

[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]

Subject: [Leica] Quantities and qualities (long, but interesting! ;)
From: Martin Howard <mvhoward@mac.com>
Date: Sun, 23 Jan 2000 18:12:18 -0500

Steve LeHuray wrote, in part:
>
> And to the best of my knowledge
> every scientific endeavor from Anthropology to cancer to film emulsion to
> snow shovels to yacht design has some baseline numbers attached to it.
>

Hehe.  Reseach is often divided into two categories: qualitative and
quantitative.  Fields of research such as anthropology and ethnography are
employed in qualitative research, where the focus in on understanding issues
such as qualitative differences, significance, and meaning.  I'm not an
anthropologist, but I've been exposed to ethnography as part of designing
interactive systems: These are used by people, and people and their actions
are not easily reduced to numbers.

You'd be hard pressed to find any respectible anthropological/ethnographical
study reduced to some baseline numbers ;)


Off-topic, and as an interesting sideline, the Western world's (or perhaps
modern civilization's even) preoccupation with quantifiable measures is an
testiment to the impact that Descarte's thinking has had.  A common reaction
today is that if it cannot be expressed in numbers, then it either doesn't
exist, or is some mushy New Age concept.

Conversly, proponents of mushy New Age concepts will often casts their ideas
in numbers, so as to lend an air of respectability to it.  Take
lens/car/yacht/hi-fi/wine/toothpick testing, for instance.  Quite often, the
test will consist of some opinionated indivudual who plays around with the
lens/car/yacht/hi-fi/wine/toothpick for half a week, assigns half a dozen
numbers to various attributes of the item under scrutiny, concludes by
adding or averaging the numbers, and then publicly proclaims that "This is a
9.6 toothpick".

They are actually doing themselves a disservice: Because they think that
numbers means objectivity and objectivity means science, and science means
respectibility, they shortcut this to think that numbers mean
respectibility.  However, philosphically, it can be argued that NOTHING is
objective (which makes for some really interesting discussions, preferably
with red wine involved, but I won't go into the details here), but even by
the commonly accepted (in Western philosophy) standards of
objectivity/subjectivity these tests cannot be considered objective.
Assigning numbers to them only clouds the real issues at hand and make it
harder for people to appreciate the value of the tests, since we end up
arguing about the numbers, rather than reading the tests and thinking about
how the tester's tastes/opinions/situation differs QUALITATIVELY from ours
(that which Mike J was so eloquently taking about).

My point?  There is no shortcut for not thinking.  While reading equipment
tests will, in a round about way, give us access to stuff we might otherwise
not learn anything about, it cannot make up our minds for us.  There are too
many variables involved, and they're not orthogonal in any case.  Even in
the most "rigorous" tests, such as Erwin's MTF evaluations, there is still
subjectivity in terms of agreeing upon what to test, how to read charts, how
equipment is calibrated, what equipment is used, and ultimately (and least
accessible) a whole, massive, shared understanding about what the world
consists of and what is real and what is not.  Want an example?  Bokeh was
nonsense twenty years ago (and still is to some).  It didn't exist.  We
still cannot reduce it to a simple number, but now there is some doubt to
the infallibility of MTF testing in capturing all important (and existing)
characteristics of a lens.

(Steve, sorry, not ranting at you, but you sparked ideas that have been
brewing for some while... ;)

M.

- -- 
Martin Howard                     | "Very funny Scotty.  Now beam down
Interactive Systems Designer      | my clothes."
email: mvhoward@mac.com           |
www: http://mvhoward.i.am/        +---------------------------------------