[
Lists Home |
Date Index |
Thread Index
]
>In the two distinct document-oriented fields of which I have some
>experience--1)
>the processing of such documents as securities orders, conditioned by
other >such
>documents as regulatory reporting requirements, and 2) the analysis of
>(mostly
>ancient) texts, chiefly formulaic and often orally composed--document
could >never
>be usefully understood as a subset of data. Documents in my experience
are >a
>precursor to data--data being one specific elaboration of
>semantics--information--from a particular process applied on a
particular
>occasion to a particular document.
>From my experience, the needs of document-oriented processing are for
more free-flowing structures, i.e. mixed content and plenty of mixing.
Documents are as I see it, structures that are human friendly to read,
and as such also human friendly to write, and data are structures that
while readable by humans are more suited to machine consumption and most
probably machine generation as well.
Documents often have specified areas that are more data like in
orientation, these upon investigation of specifications relating to our
document type will most probably turn out to be metadata.
I got into this game because I prefer Document-oriented fields, and am
very interested in such. Somewhere deep inside, I think I might actually
hate data. ;)
|
