ONT Re: Data Models, Ontologies, Logic
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
Note 13
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
Had I been born into my surly office, and never left it since birth,
I might have some funny ideas about what data is and where it comes
from, its ontology and its ontogeny, if you will, but as it happens
there were times o' plenty -- well, really, o' scarcity -- when I
designed my own interview instruments, collected the data, sorted,
carded, spindled, folded, and data-reduced the entire harvest by
the sweat my brow and me and my family's sheer manual labor, and
carried out a full-fledged statistical analysis with nothing more
than pencil, paper, and the odd neuron or three.
So I will not forget that the most important thing about the data
is how it refers to the objective world from which it is gathered,
in the manner of signs that are abstractedly sampled from objects,
by virtue of which sampling they denote the vaster reality beyond.
It's always been declarative, always been denotative, all the way down.
Here are some randomly sampled thoughts that
come to mind as I review the present example.
Though we may later think on it as "raw data", what you see in that dataset
is nothing at all like raw data, properly speaking. The closest thing that
we have to raw data in this particular paradigm would be the video tapes of
the family interaction observational sessions. The codebook is the outcome
of a decade's development within a specialized subcommunity of inquiry, and
so it is a type of cultural reticle, scope, template, in its own right, all
too often preveiling as an irreflective transparency projected on the scene.
What goes into the constitution of the codebook is a mixture of theoretical
bias and personal perspective, a hefty dose of practical experience, not to
mention a bit of factor analysis, multiple regression, and reliability work.
That's about as raw as a code ever gets.
So you might view the dataset, with or without its interpretive coda,
as a particular genre of text, that manifests a specialized style of
elliptic, hyperbolic, parabolic description with regard to the world
that all the world calls "the world", and even if you do not find in
this text a thriller very much to your taste, I have known a species
of data-hound who'd already be salivating over its suspended mystery
and tearing into its matter stem-and-leaf, like the well-conditioned
dog of the proverbial sort.
Jon Awbrey
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o