0 RUNNING HEAD Investigating MODS Eleftherion Carr Investigating MODS

0 running head: investigating mods eleftherion carr investigating mods melissa eleftherion carr a-3 libr 202 - 23 san

0
RUNNING HEAD: Investigating MODS Eleftherion Carr
Investigating MODS
Melissa Eleftherion Carr
A-3
Libr 202 - 23
San Jose State University
May 15, 2010
Much like the issue of ambiguity of natural language use in metadata
design, there is somewhat conflicting information about how and why
MODS (Metadata Object Description Schema) was created. According to
the Library of Congress’ official MODS website, it is an “XML markup
for selected metadata from MARC21 records as well as original resource
description.” (Library of Congress, 1999). It is categorized under the
Resource Description Format header. According to Alemneh, MODS was
“originally designed for library use but may be used for other
applications.” (Alemneh, 2007, slide 4). In his prefatory statement to
Rebecca Guenther’s article introducing MODS, Martin Dillon states that
MODS was “developed as an initiative of the Digital Library Federation
with help from the library and information science communities.”
(Guenther, 2003, p. 138) However, just further down the same page,
Guenther writes, “ Library of Congress’ Network Development and MARC
Standards Office…developed…the Metadata Object Description Schema
(MODS), which includes a subset of MARC elements.” (Guenther, 2003, p.
138). By revisiting a different page of LC’s website: MODS: uses and
features, one will learn that Guenther’s information is indeed
correct. One further point, though, is while Dillon writes, “MODS was
designed to be used with The Metadata Encoding and Transmission
Standard, (METS)” (p.138, emphasis mine), Guenther states “the project
is experimenting with METS for packaging the digital object and its
metadata.” (p. 148, emphasis mine). Metadata standards, by their
nature, are valuable because they provide a structure that helps
mitigate semantic issues. It is curious then, and somewhat amusing,
that these semantic variations exist to define a metadata standard
that seeks to provide solutions for these semantic issues.
There are, however, characteristics of MODS that these various sources
agree upon. As an XML-based descriptive metadata standard, MODS was
essentially created to assume consistent bibliographic control over an
ever-evolving, perpetually growing network of electronic information
resources. Due to its consistency, and high degree of
interoperability, it is believed to have unique capabilities for
sharing information with several other metadata standards, namely MARC
and MARC-21. In fact, it is considered to provide very good
round-tripability with MARC records due to its origin as a MARC
derivative. It is also considered to be simpler than MARC-21 but
richer than Dublin Core while retaining variable length fields, due to
its XML compatibility. Its’ language-based tags also enhance its
attractiveness to users by appearing more friendly than MARC or MARC
21’s numeric-based tags. This simple capability seems to enable MODS
to suit any English-speaking person’s needs. Currently, MODS is being
marketed as a highly shareable and re-purposeable metadata standard,
which has the capability to crosswalk with numerous standards,
including EAD and simple Dublin Core.
MODS’ schema appears to be clean and concise. There are twenty defined
top-level elements in version 3.3, although many of them contain
several sub-elements. The hierarchical order is as follows: titleInfo,
name, typeofResource, genre, originInfo, language,
physicalDescription, abstract, tableofContents, targetAudience, note,
subject, classification, relatedItem, identifier, location,
accessCondition, part, extension, and recordInfo. For the relatedItem,
any MODS element may be used as a subelement. According to Guenther, “this
substructure has proven particularly useful in digital library
projects, where there is a need to describe the item as a whole as
well as its subparts using relatedItem with type=“constituent” (e.g. a
CD with its constituent songs by different artists on separate
tracks).” (Guenther, 2003, p.147). One top level element worth
mentioning, upon comparison with Dublin Core, is name="name" type="nameType"/> (name). DC uses both the terms: Creator
and Contributor in its comparable element that has been lauded as
“inadequate” and “broad.” (Guenther, 2003, p. 43). However, the
simplicity of MODS’ element “name” allows for either personal or
corporate names, and does not seek distinction.
Both MODS and METS were first used at LC in 2002, for their Digital
Audio-Visual Preservation Prototyping Project. They used them both
again in 2003, yet this time was their first in production. And, later
in 2003, Veteran’s History Project database, a MINERVA project was
released, which also used MODS. In 2006, MODS was employed by LC for
the National Digital Newspaper Project as a repository submission
package (Library of Congress, 2009).
According to LC’s MODS Implementation Registry, there are currently
thirty-one implementers of MODS. In this paper, I will discuss the
following projects and how MODS is being used: TDL Repository (Texas
Digital Library), U.C. Berkeley Library Archival, Rare and Fragile
Collections (California Digital Library/Calisphere/Online Archive of
California), and DLF Aquifer Initiative (Digital Library Federation).
The Texas Digital Library created an application profile in December
2005, converted their records from DSpace Dublin Core to MODS in
January 2006, and premiered online in February 2006. They currently
use version 3.1.
The TDL Repository is composed of collections digitally archived by
the five Association of Research Libraries (ARL) universities in the
state of Texas. The TDL Repository serves to preserve and promote the
research output of Texas, including electronic theses and
dissertations (ETDs). TDL is using MODS as the common descriptive
metadata syntax for ETDs. (Library of Congress, 2010, MODS
Implementation Registry)
Although TDL appears confident in their use of MODS, there also
appears to be an issue of time consumption for their staff. In their Descriptive
Metadata Guidelines for Electronic Theses and Dissertations, Version
1.0, the TDL Metadata Working Group note,“ because DSpace requires a
flat, Dublin Core‐like schema, TDL partners should be aware that at
present, MODS cannot be directly ingested into the TDL ETD repository
by any means other than the Thesis and Dissertation Submittal System.”
(TDL Metadata Working Group, 2008). They recommend mapping techniques
for preserving the richness of the data, and provide detailed notes
and tables. One might argue that the need for such a detailed document
is in itself an indication that the use of MODS presents difficulties
for TDL’s staff. However, this attention to preservation does not seem
to affect ETD creators/users who ultimately benefit from the richness
of MODS.
University of California Berkeley Library (UCB) has a grant-supported
ongoing project wherein they “have digitized and continue to digitize
portions of its archival collections and selected rare and fragile
materials. The digitized materials are exported as METS objects with
MODS encoded descriptive metadata. Most of these METS objects are then
ingested into the California Digital Library\'s Online Archive of
California and Calisphere.” (Library of Congress, 2010). Maintained by
California Digital Library (CDL), Calisphere and OAC are available for
public use and aim to provide “access to more than 200,000 primary
sources such as photographs, documents, newspapers, political
cartoons, works of art, diaries, transcribed oral histories, and other
cultural artifacts.” (CDL, 2010) They currently use MODS version 3.2.
The CDL strongly supports the assertion that Dublin Core does not
provide enough encoding granularity. The CDL therefore prefers that
descriptive metadata is encoded in a richer format, such as MODS.
Institutions should use qualified Dublin Core only in cases where MODS
is not locally supported. (CDL GDO, 2010). As is the case with TDL, it
is once again Dublin Core that presents crosswalk issues with MODS. In
their California Digital Library Guidelines for Digital Objects (CDL
GDO), they outline mapping techniques for numerous attributes
including date, title, creator, and language. However, both TDL and
CDL maintain that the quality of MODS’ data far outweighs the
requisite time involved.
The DLF Aquifer is an initiative of the Digital Library Federation.
Their primary goal is “to promote effective use of distributed digital
library content for teaching, learning, and research in the area of
American culture and life. ” (Shreeves, 2007). Despite the fact that
many participating libraries were not using MODS, the DLF Aquifer
Initiative decided to convert in 2005, mostly due to “lack of semantic
complexity” in Dublin Core (Shreeves, 2007). As their priority is
sharing and harvesting metadata, this may have been considered to be
quite risky. One has to admire this kind of integrity and faith in
their work. They believed that they would be setting standards for
best practices in sharing metadata because “requiring MODS records
would start the DLF Aquifer Initiative with rich, semantically complex
records and would avoid the already well-documented challenges of
working with unqualified DC.” (DLF Aquifer Initiative, 2006). After
releasing the implementation guidelines to the public for two months
for review and comment, they received two comments noting just two
major issues (DLF Aquifer Initiative, 2006). The first concerned “the
requirement of one and only one pair for the main
portion of the record (i.e. multiple pairs could be
used within the element).” The second concerned “how and
where to describe the digital surrogate and the analog original, or
how to present information about the content of a resource and its
carrier(s).” (2006). Both of these issues received due attention, and
appeared to have been resolved without further issue. Due to the
prevalence of simple DC among many libraries in the DLF and the
“simple Dublin Core requirement of the OAI protocol”, the Aquifer
Initiative have suggested mappings for crosswalking MODS to DC. They
maintain, though that simple DC is not recommended as “the primary
metadata format.” (DLF Aquifer Initiative, 2006). Like TDL and CDL,
the DLF are supporters of MODS and its myriad capabilities.
Although it was originally designed for “a variety of purposes, and
particularly for library applications” (Library of Congress, 2009),
one foresees many possibilities for MODS. There are many proponents of
MODS that claim it as a future replacement for MARC and Dublin Core
(DC) due to its richness and superior granularity. Moreover, as MODS
was designed to work with XML, it is currently the only standard agile
and versatile enough to be considered as a replacement for MARC.
References
Alemneh, D. G. (2007). An introduction to MODS: the metadata object
description schema. [Powerpoint slides] Retrieved May 9, 2010 from
http://www.library.unt.edu/digitalprojects/tech-talks/mods/
California Digital Library. (2010, March 12). Calisphere. Retrieved
May 4, 2010 from http://www.cdlib.org/services/dsc/calisphere/
California Digital Library. (2010, January). CDL guidelines for
digital objects, version 2.0. [PDF] Retrieved May 9, 2010 from
http://www.cdlib.org/services/dsc/contribute/docs/GDO.pdf
Cundiff, M. (2009). Using and to create XML
standards-based digital library applications. [Powerpoint slides].
Retrieved May 15, 2010 from
http://www.loc.gov/standards/mods/presentations/mets-mods-morgan-ala07/pages/mets-mods-morgan-Slide07_jpg.htm
Guenther, R. S. (2003). MODS: the metadata object description schema.
Libraries & the academy. 3(1) pp. 137-151. Retrieved May 9, 2010 from
Academic Search Premier database.
Digital Library Federation Aquifer Initiative. (2006). Digital library
federation aquifer project update. D-Lib Magazine 12(5) Retrieved May
15, 2010 from http://www.dlib.org/dlib/may06/kott/05kott.html
Library of Congress. (1999, November 25). Standards at the library of
congress. Retrieved May 14, 2010 from http://www.loc.gov/standards/
Library of Congress. (2009, July 28). MODS: uses and features.
Retrieved May 15, 2010 from
http://www.loc.gov/standards/mods/mods-overview.html
Library of Congress. (2010, May 10). MODS implementation registry.
Retrieved May 10, 2010 from
http://www.loc.gov/standards/mods/registry.php
McCallum, S. H. (2004). An introduction to the metadata object
description schema: MODS. Library HiTech. (22)1 pp. 82-88. Retrieved
May 9, 2010 from ABI/Inform Global database.
Shreeves, S. (2007). Creating Rich Shareable Metadata: the DLF Aquifer
MODS implementation guidelines. [Powerpoint slides] Retrieved May 15,
2010 from
http://www.loc.gov/standards/mods/presentations/shreeves-ala07/
TDL Metadata Working Group (2008). Texas Digital Library Descriptive
Metadata Guidelines for Electronic Theses and Dissertations, Version
1.0. [PDF] Retrieved May 15, 2010 from http://www.tdl.org