16
Information Classification 101 –
Part One: a Primer on Metadata
As the rate of information growth accelerates, information classification
continues to increase in importance as a tool to manage that growth.
By, Bob Rogers Chair, SNIA DMF Information Lifecycle Management
Initiative.
DMF SNIA
SNIA
Paradoxically, a large percentage of IT people see infor- December 2008, when it required the adoption of the
mation classification as difficult, labor intensive, and eXtensible Business Reporting Language (XBRL). XBRL
error prone. This series of articles is intended to clari- is an example of an international standard for descrip-
fy some of the issues with the process, identify several tive metadata used to communicate business and
of the techniques, and show how each technique financial information. Descriptive metadata is not the
applies to the overall objective of classifying or catego- only type; NISO has defined other forms including
rizing information objects within the data center. “structural” and “administrative” metadata.
Perhaps one of the more difficult facts to accept is that
there is no simple “universal” method of classifying Structural metadata is an interesting concept because it
information. The needs of information stakeholders defines the relationships of a compound object. Most
form the basis for the methodology used to get the job databases are a compound object made up of data
done. Although the Information classification process files, tables, indices, logs, and other entities. (Most
can take many different forms, the one attribute com- storage people figured out the compound nature of a
mon to every information classification methodology is database the hard way, when something was done to
the necessity for metadata. What is metadata? one physical component and the whole thing quit
working.) Structural metadata may include more than
The term is used frequently and often without much just physical relationships; for example, the relation-
consideration for its implications, as in “data about ships between bits of knowledge represented by struc-
data.” The National Information Standards tural metadata might produce important information
Organization (NISO) defines metadata as “structured without the need to access the data that it describes.
information that describes, explains, locates, or other- Storage professionals are much more familiar with
wise makes it easier to retrieve, use, or manage an “Administrative metadata.” NISO defines it as “infor-
information resource.” The NISO definition introduces mation to help manage a resource such as when and
the important concept that metadata has a purpose
1
. how it was created, file type, and other technical infor-
mation.”
There are at least two types of metadata used for
information classification; the first type is the criteria or File system metadata is a good example of administra-
set of rules to apply to the information source; for tive metadata. Administrative metadata is not limited
example, search arguments. The second type of meta- to that produced by a system; it could also be the
data is an index, or an extract, or a summary, of the metadata produced by a person such as authorship or
information source. Both of these types of metadata other document properties. Yet another example of
are “descriptive” because they describe a resource for administrative metadata describes the rights of a user
discovery or identification. A major step forward in or organization to access information.
business information classification was announced by
the Securities and Exchange Commission (SEC) in Recall that this article began with the assertion that
FEB 09 WWW.SNSEUROPE.COM
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36