Understanding Metadata and Metadata Schemes, Part I, Intellectual Foundations/ Jane Greenberg

Jane Greenberg is Associate Professor for the School of Information and Library Science at the University of North Carolina at Chapel Hill.

[In a previous blog entry, I discuss my ire with the term “metadata.” In lieu, I will use “supradata” unless I am quoting directly. I know that this is pedantic, nonstandard, and therefore presents an irritation; I humbly beg the reader’s forgiveness in advance.]

Greenberg and Kant
Supradata represent a philosophically rich subject for Greenberg because she is concerned with systematizing them. She adopts a Kantian stance as a pragmatic way to approach the subject, and it is clear Dr. Greenberg has done a lot of thinking about it. I consider it a privilege to be exposed to her thoughts. (It is stuff like this that makes me wonder – what good have I done in life to be blessed with the fruit our professor’s labor in selecting Greenberg’s work? Surely, I will need to reply to that question another time.)

Greenberg’s Functional Classification
Greenberg uses a functional classification, listing seven examples of prototypical supradata (my examples in parentheses):

1. Identification/description (e.g., author, subject, title…/ about global warming…)
2. Administrative (e.g., these data may be downloaded at… )
3. Terms and Conditions (legal notices… )
4. Content rating (e.g., PG-13, not suitable for minors…)
5. Provenance (e.g., ice-core data from Prudhoe Bay glacier…)
6. Linkage/relationship (e.g., similar to Millikan’s findings…)
7. Structural (e.g., requires Adobe Acrobat Reader…)

I am not sure why the above classification does not differentiate origination from “aboutness” in Item 1 because that seems an important distinction. Greenberg observes that “…literature does not reveal a universally accepted definition for metadata scheme – unlike the standard definition of ‘data about data’ for metadata.” And she is right, of course.

A Schematic for Schema
This leads me to ask, is there a normative model for classification of supradata? If supradata are “data about data” then shouldn’t supradata classifications comprise the taxonomy of “data about data,” i.e., structural relations among features.

I know that Aristotelian (or even Linnaean) taxonomy is becoming passé and is essentially opposed to a Kantian philosophical construct. I think the most disturbing part of Aristotelian taxonomy to some is the underlying premise of a perfect referent. These are Aristotelian and Platonic ideals, and Kant has supposedly done away with them – but in my opinion, prematurely.

Am I blind? Has my hammer simply found yet another nail? Quite possibly, but this only underscores my maxim – it is all about worldview (you may say paradigm if you must); I am enlightened by it (or blinded, as you may prefer) as are we all. I think Greenberg has done an outstanding job in systematizing and codifying supradata schema. Perhaps she would object to framing things in a normative way, though she references normative standards such as the Paris Principles and Cutter’s objectives. However, she adopts a very useful and pragmatic framework approach (MODAL), which another library science student has summarized masterfully in her blog.

But Origination/ Identification/ Administration/ Notification/ Ratification / Provenance/ Relation/ Mechanization… I think they fit comfortably within a normative scheme. I suppose I am hope(lessly/fully) bent on exploring the mind of God and thinking His thoughts after Him, but it has proven most useful historically; because, ultimately, there is nothing more pragmatic than truth.