Scott Carlson | 15 Sep 20:51 2014

Job Announcement: Database & Metadata Management Coordinator - Houston, TX

-- Apologies for Cross-Posting --

Rice University is seeking a creative, dynamic, service-oriented Database &
Metadata Management Coordinator. The successful candidate will provide
leadership in performing and managing bibliographic data quality control,
database maintenance and problem resolution. The position reports to the
Head of Cataloging and Metadata Services in Technical Services.

Responsibilities:  Coordinates authority control of both MARC and non-MARC
metadata to ensure that they meet local and national standards. Develops new
and effective approaches to quality control, consulting with various
stakeholders as appropriate. Serves as a liaison to authority and database
enrichment vendors and manages the loading and quality control for
vendor-supplied data. Collaborates with other staff to provide leadership in
transitioning the department to a post-MARC communication format as we
continue to contribute metadata to the local institutional repository and
proactively explore linked data, the semantic Web and BIBFRAME. Assists in
and/or manages special projects undertaken by the Department as assigned by
the Department Head. Participates in departmental and library-wide
committees and work groups established to further various aspects of the
Fondren Library mission. May lead project teams. Assists in other cataloging
activities as needed.

Required Qualifications:
-ALA-accredited Master's degree in Library Science
-two years of successful related experience in quality & authority control
of MARC and non-MARC metadata
-some successful project management experience, preferably in an academic
-Excellent oral and written communication skills.
(Continue reading)

Meehleib, Tracy | 11 Aug 21:49 2014

MARCXML to MODS 3.5 -- XSLT 1.0 (Revision 1.101) Announcement

MARCXML to MODS 3.5 -- XSLT 1.0 (Revision 1.101) Announcement

The Library of Congress' MARCXML to MODS 3.5 (XSLT 1.0 Revision 1.101) is now available via the Library of Congress' MODS Web site and at < >--it incorporates edits made in response to comments received since the release of Revision 1.96.

The MODS 3.5 XSLT is based on the MARC to MODS 3.5 mapping made available by the Library of Congress and last updated in May of 2014 <>.

Please note that content standards and practices using MARC vary among institutions, and sometimes within an institution over time or between types of collections or materials. No one stylesheet can deal appropriately with all variations in the use of MARC. The mapping decisions underlying the changes we have made, as well as the decisions underlying the LC stylesheet should be tested on sample records to determine whether the results are as desired or whether the institution needs to modify the stylesheet for a particular collection of records.

General questions about the mapping may be directed to the MODS Editorial Committee members via Specific questions about the stylesheet may be addressed to Tracy Meehleib at .

Thank you,


Tracy Meehleib

Network Development and MARC Standards Office
Library of Congress
101 Independence Ave SE

Washington, DC 20540-4402
+1 202 707 0121 (voice)
+1 202 707 0115 (fax)


Rebecca Guenther | 11 Aug 21:11 2014

AMIA cataloging and metadata workshop

This workshop might be of interest to those of you managing moving image (analog and digital) materials. It’s being held as a pre-conference workshop at the Association of Moving Image Archivists (AMIA) conference, but attendees can register separately. This year the conference has a special stream on “digital” aspects of preserving and providing access to digital AV content. More info on the conference in general:


AMIA Cataloging & Metadata Workshop | October 7-8, 2014 | Savannah, Georgia

The 2014 edition of this bi-annual workshop includes new content and offers attendees—past and future alike—a workshop that emphasizes practical implementation through the use of hands-on exercises and addresses how the attendee can concretely integrate new models for description into their current workflows and environments.    The two-day program moves the attendee from concept to implementation starting with cataloging principles and metadata concepts that form the basis for describing and providing access to moving images in an evolving media and media technology environment. The workshop includes information about the role of cataloging for analog and digital asset management; the value, purpose and application of metadata and cataloging standards; management of resources through their life cycles; descriptive, structural, and administrative metadata (including rights and preservation metadata); and, data models and data mapping. Dynamic presentations encompass film, video, digital, and broadcast materials and include interactive exercises to put cataloging and metadata concepts directly into practice. 


Please excuse duplication of this message.


Melanie Wacker | 8 Aug 22:27 2014


the MODS Editorial Committee would like to follow up on a proposal to introduce some changes to the MODS <hierarchicalGeographic> element.  While there are no specific plans for a MODS version 4 at this point, it would be a major overhaul of MODS. However, several useful changes to the <hierarchicalGeographic> element could be introduced in MODS 3.6.

A previous proposal was sent out for discussion on Dec. 27, 2013.
This proposal can be found here:


The MODS EC revised the proposal based on the feedback received and is now soliciting comments on the new proposal, see below.
We would like to hear from you by Sept. 5, 2014

Melanie Wacker

on behalf of the MODS EC


MODS HierarchicalGeographic Proposal

1      Specific Proposal 

1.1    State and Province
<province> will be deprecated;  <state> will be re-defined:
<state>-- Includes first order political jurisdictions under countries, such as states, provinces, cantons, Länder, etc. regardless of what they are called in the particular country.

See Commentary and Examples:  1.

1.2    New Attributes to Indicate Place Types

Attributes <at> areaType, <at> regionType, and <at> sectionType will be defined for elements <area>, <region>, and <citySection>  respectively.  These would of course be optional.

See Commentary and Examples:  2.


1.3    Indication of hierarchical level


Attribute <at> level will be defined for all place type elements to indicate hierarchical level. It would of course be optional.

See Commentary and Examples:  3.

1.4      Authority

The authority attributeGroup (authority, authorityURI, and valueURI)  will be added to all place type elements.

1.5    Places that no longer exits

Attribute <at> period will be defined. Its presence will indicate that the described entity once existed but no longer exists.  Its value would be a hint of when it existed (it could simply be a date).

            See Commentary and Examples:  4.


2      Commentary and Examples


  1. The way MODS currently handles first level political jurisdictions -- i.e. first level below the country -- is not consistent. It was brought out that in the MARC 662 there is a subfield called "first order political jurisdiction". In MODS, both <state> and <province> are used for first order political jurisdictions depending upon what they're called in the particular country-- but the guidelines say the following:

<state> – Includes first order political divisions called states within a country, e.g. in U.S., Argentina, Italy. Use also for France département.

<province> – Includes first order political divisions called provinces within a country, e.g. in Canada.

So first order political divisions mostly go under <state> unless they're called <province>. And we're silent on what they call them in other countries.

The committee thinks it desirable to put all first order political divisions under one element. The term "first order political division" or "first order political jurisdiction" is rather unwieldy. So the proposal is to deprecate province and define <state> as all first order political jurisdictions.

  1. This replaces the earlier proposal to define new element <placeOther> and attribute <at> otherType, to accommodate other types of places that don't have their own element.  Instead,  group them together under <area>, <region> or  <citySection> as appropriate (area is used for non-jurisdictional places, region for jurisdictional,  and <citySection> may be used for either ) and indicate the place type with the corresponding attribute, <at> areaType,   <at> regionType, or <at> sectionType.

<citySection> for example is currently defined as:

<citySection>-- Name of a smaller unit within a populated place, e.g., neighborhoods, parks, or streets

Thus, citySection is a broad term that can be used without the <at> sectionType attribute, but if you want to designate a specific type of city section you  could say

<citySection sectionType="neighborhood" >

for formally established neighborhoods, or

            <citySection sectionType="street>".

Note that the use of "city" here doesn't preclude  towns.

<area> could take an <at> areaType attribute to accommodate some of those areas that have been suggested-- national parks, rivers, Indian reservation, etc.



            <country>United States</country>

            <state>Rhode Island</state>


            <citySection citySectiontype="neighborhood">Blackstone</citySection>


  1. When the level of elementA is less than the level of element, then elementA is higher in the hierarchy than element. It is recommended that levels begin with 1 and are consecutive, however this is not mandated. Levels could, for example, be 3 , 5, and 9.

There is no need for <at> level  for simple cases like the following where the hierarchy is easily inferred:

            <country>United States</country>

            <state>Rhode Island</state>


 But there are a few cases where <at> level would be useful:

  • The actual hierarchy of the existing place types for hierarchicalGeographic isn't always clear, and in some cases depends on the context.

  • It will be useful to be able to indicate that two places have the same level.

  • In cases where the same place type is supplied more than once, their levels relative to one-another are not always clear.



 To elaborate of the latter two points:

 Consider the following example, where Blackstone is a sub-neighborhood of East Side, in Providence.

            <country>United States</country>

            <state>Rhode Island</state>


         <citySection citySectiontype="neighborhood" level="1">East Side</ citySection >

      < citySection citySectiontype="neighborhood" level="2">Blackstone</ citySection >

 Another example, consider Massachusetts Avenue which runs through Lincoln Park within Capitol Hill in Washington DC. Say you want to identify that section of Mass Ave  within the park: 

           <state>District of Columbia</state>


            <citySection citySectionType ="neighborhood" level="1">Capitol Hill</ citySection>

            < citySection citySectionType =”park”  level="2">Lincoln Park</ citySection>    

          <citySection citySectionType =”street” level="2">Massachusetts Avenue</citySection>

Whenever the level is the same for two places the intersection is indicated. This example indicates the intersection of the two level 5 entities.  


  1. The following example indicates the portion of the Oregon Trail within Idaho that existed during the gold rush: 

           <state level="2">Idaho</state>

            <area  level="3"type="trail" period="gold rush">Oregon Trail</area>

Another Example:


            <country period=”1945-1990”>East Germany</country>
Elizabeth Post | 4 Aug 15:36 2014

METS Workshop Invitation

All are invited to participate and provide feedback to the METS Editorial Board on a new draft version of the METS schema, METS 2.0, at a workshop being held in London as part of Digital Libraries 2014, an event organized by the ACM/IEEE Joint Conference on Digital Libraries (JCDL 2014) and the International Conference on Theory and Practice of Digital Libraries (TPDL 2014). 

The Metadata Encoding and Transmission Standard (METS) 1.x schema has an established community of users many of whom are moving toward the use of newer technologies such as those of the Semantic Web and linked data for their digital content and metadata. In this workshop, entitled METS Now, and Then… Discussions of Current and Future Data Models, participants will develop an understanding of the data models underlying some canonical uses of the existing METS schema as illustrated by a panel of experienced METS implementers.  The explication of data models will provide a contextual basis for the description of a next generation METS (2.0) data model.  Participants will be invited to participate in the refinement of the METS 2.0 data model being developed by the METS Editorial Board, and discuss options for serialization of the data model.

The workshop will take place over two half days:  Thursday, September 11 (afternoon) andFriday, September 12.  For more information and registration, please visit the Digital Libraries 2014 site or contact Betsy Post (

Please note that the last day for discounted early registration is Monday, August 11.

Meehleib, Tracy | 30 Jul 21:10 2014

LC NDMSO survey monkey XSLT 1.0 vs 2.0

Dear MODS Users,


The Library of Congress would like to better understand how its MODS stylesheets are being used. If you use any of LC’s MODS-related  XSLTs to convert records, and haven’t already responded to this survey, please take a moment to fill it out:  

< >.  


Thank you!




Tracy Meehleib

Network Development and MARC Standards Office
Library of Congress
101 Independence Ave SE
Washington, DC 20540-4402
+1 202 707 0121 (voice)
+1 202 707 0115 (fax)



McCallum, Sally | 29 Jul 23:10 2014

AV study posted

Just posted on the BIBFRAME (BF) web site is a study carried out by Audiovisual Preservation Solutions for the Library of Congress concerning an appropriate model for AV material and relating that to the BF model:  BIBFRAME AV Modeling Study: Defining a Flexible Model for Description of Audiovisual Resources.   It discusses the special characteristics of AV material that make it different from textual and other media and makes some general recommendations.  With the increasing impact of AV as information resources and the need to preserve material in those media, special attention to it is appropriate as we develop BIBFRAME.


This report was carried out in close consultation with the Library of Congress’s National Audiovisual Conservation Center (NAVCC) staff in Culpeper, Virginia, a vast archive that preserves and serves AV resources.  The report examines a number of proposed community models such as FRBR/RDA, FIAF, OLAC, indecs, Variations, PBCore, EBCore and others that inform the analysis.    The recommendations will become considerations for the BF model development as there are various ways the special attributes of these media could be accommodated.


The report is composed of a base document (the analysis, AV model, and recommendations) with appendices that give more detail about situations encountered with AV material.


As usual, comments, concerns, and discussion are encouraged via the BIBFRAME listserv (see the BIBFRAME home page/contacts) or direct to .





Sally H. McCallum

Chief, Network Development and MARC Standards Office

Library of Congress, 101 Independence Ave., SE

Washington, DC 20540  USA

Tel: 1-202-707-5119 – Fax 1-202-707-0115




Lina Bountouri | 28 Jul 14:47 2014

Final Call for Papers: MTSR 2014 - Special Track on Metadata & Semantics for Cultural Collections & Applications

Apologies for cross-posting



Part of the 8th International Conference on Metadata and Semantic Research (MTSR 2014), 27 - 29 November 2014, Karlsruhe, Germany

Submission deadline: New Deadline! July 31st, 2014

Proceedings will be published in Springer CCIS series



Cultural Heritage collections are essential knowledge infrastructures that provide a solid representation of the historical background of human communities. These knowledge infrastructures are constructed from and integrate cultural information derived from diverse memory institutions, mainly museums, archives and libraries. Each individual community has spent a lot of effort in order to develop, support and promote its own metadata as tools for the description and dissemination of cultural information, mainly related to its particular resources and use.

The exposure of cultural information into the Semantic Web makes clear that metadata have to be accurate and deeply interpretable in the semantic level. Conceptual Reference Models could facilitate these procedures since they constitute conceptualizations - according to the specific point of view of a memory institution or its particular community - providing at the same time the context for interpreting the respecting metadata to their domain of discourse. CRMs have also been viewed as global schemata in order to map different metadata specifications. At the same time, there are also inter-domain efforts targeted to semantically align data (research data, educational data, public sector information etc.) to cultural information. The management of the cultural information provides challenges associated with (i) metadata modeling, specification, standardization, extraction, evaluation, mapping, integration and effective use, (ii) knowledge representation as conceptualization to provide the context for unambiguously interpreting metadata, and (iii) information integration from different contexts for the provision of integrated access and advanced services to the users.

The aim of this Special Track is to maintain a dialogue where researchers and practitioners working on all the aspects of the cultural information will come together and exchange ideas about open issues in all stages of the metadata life cycle. The track also welcomes works for the connection and interlinking of the Cultural Heritage metadata to any other dataset published in the Semantic Web universe.



The papers in this special track should be original and of high quality, addressing issues in areas such as:

  • Cultural heritage metadata models, standards, interoperability, mappings and integration
  • Automated metadata extraction
  • Ontologies and knowledge representation for the cultural heritage domain
  • Extracting semantics, entities, and patterns from Cultural Heritage collections
  • Collection models and item - collection relationships representation
  • Collection - level metadata modeling and management
  • Linked open data approaches for the cultural heritage domain
  • Composite content-discovery and management of components and interrelationships
  • Large volume content management - high resolution image data sets
  • 3D models-indexing, storage and retrieval approaches
  • Federation of repositories/data infrastructures
  • Integration of intra or inter disciplinary heterogeneous resources
  • Infrastructures for sharing content
  • Digital Curation workflows and models
  • Preservation metadata for cultural heritage digital objects
  • Metadata quality metrics
  • Case studies



Authors can submit either full papers (12 pages) or short papers (6 pages). Submitted papers have to follow the LNCS proceedings formatting style and guidelines.

The submitted papers will undergo the same peer review as the submissions for MTSR 2014 and accepted contributions will be published in the MTSR 2014 proceedings (Springer CCIS series). Authors of accepted papers will be asked to register to the Conference and present their work.

Authors of the best papers will be invited to submit extended and revised versions of their papers for possible publication in selected international journals, including the International Journal of Metadata, Semantics and Ontologies (Inderscience), and Program: Electronic library and information systems (Emerald) (list incomplete).

More information on submission can be found at the MTSR 2014 call for papers web page.



New Deadline! July 31st, 2014: paper submission

New Deadline! August 25th, 2014: acceptance/rejection notification

August 31th, 2014: camera-ready papers due

November 27th - 29th, 2014: MTSR 2014





  • Trond Aalberg, Norwegian University of Science and Technology (NTNU), Norway
  • Anila Angjeli, Bibliotheque National de France, France
  • Karin Bredenberg, The National Archives of Sweden, Sweden
  • Costis Dallas, Faculty of Information, University of Toronto, Canada
  • Manolis Gergatsoulis, Ionian University, Greece
  • Antoine Isaac, Vrije Universiteit Amsterdam, Netherlands
  • Sarantos Kapidakis, Ionian University, Greece
  • Irene Lourdi, Libraries Computer Centre, National & Kapodistrian University of Athens, Greece
  • Maria Emilia Masci, Scuola Normale Superiore di Pisa, Italy
  • Carlo Meghini, National Research Council of Italy (ISTI-CNR), Italy
  • Christian-Emile Ore, University of Oslo, Norway
  • Christos Papatheodorou, Ionian University and Digital Curation Unit, IMIS, Athena RC, Greece
  • Daniel Pitti, Institute for Advanced Technology in the Humanities, University of Virginia, USA
  • Stephen Stead, Paveprime Ltd., UK
  • Chrisa Tsinaraki, Joint Research Centre, European Commission, Italy
  • Andreas Vlachidis, Faculty of Computing, Engineering and Science, University of South Wales, UK
  • Katherine Wisser, Graduate School of Library and Information Science, Simmons College, USA
  • Maja Žumer, University of Ljubljana, Slovenia

Melanie Wacker | 25 Jul 22:33 2014

MODS relatedItem to associate datasets with research publications

The MODS Editorial Committee has considered the proposal to add a relatedItem type value for expressing the relationship between a research publication and the data set(s) that it is based on. A comparable element describing this relationship is included in MARC 21, 786 (Data Source Entry). However, it was also pointed out on the listserv that only including a relatedItem type for datasets may be too narrow to address the larger issue of different types of accompanying materials deposited in repositories and that an approach allowing for the inclusion of external vocabularies may be preferable.
The entire listserv discussion can be found here:

The  MODS Editorial Committee feels that new type values for relatedItem should be added only when they are strongly justified. Instead, we would like to introduce a mechanism whereby type values from external vocabularies can be supplied.   We propose the addition of 4 new attributes:

  • otherType

  • otherTypeAuth

  • otherTypeAuthURI

  • otherTypeURI

<at> otherType may be supplied alone, in which case it is an uncontrolled type.
If <at> otherType is supplied along with <at> otherTypeAuth, then it is a controlled type where <at> otherTypeAuth is its authority.
<at> otherTypeAuthURI may be used to supply a URI uniquely identifying the vocabulary from which the controlled type has been selected.
<at> otherTypeURI may be used to supply a URI uniquely identifying the controlled type value from a vocabulary.

The MODS Editorial Committee solicits comments on this proposal. We'd like to receive feedback by August 15.

Melanie Wacker
on behalf of the MODS EC
Meehleib, Tracy | 24 Jul 20:26 2014

Library of Congress NDMSO MODS XSLT 1.0 vs. XSLT 2.0 Survey

Dear MODS Users,


The Library of Congress would like to better understand how its MODS stylesheets are being used. If you use any of LC’s MODS-related  XSLTs to convert records, please take a moment to fill out our survey here:  < >.


Thank you!




Tracy Meehleib

Network Development and MARC Standards Office
Library of Congress
101 Independence Ave SE
Washington, DC 20540-4402
+1 202 707 0121 (voice)
+1 202 707 0115 (fax)


Kevin S. Clarke | 22 Jul 20:20 2014

MODS to DC XSLT (MODS3-5_DC_XSLT1-0.xsl) patch


I'm not sure if this is the right place to report this, but I believe I've found something in the MODS to DC XSLT (MODS3-5_DC_XSLT1-
0.xsl) that could be fixed.

We were getting empty dc:subject elements in our XML from the following template:

<xsl:template match="mods:subject[mods:topic | mods:name | mods:occupation | mods:geographic |
mods:hierarchicalGeographic | 
mods:cartographics | mods:temporal] ">
    <xsl:for-each select="mods:topic | mods:occupation">
      <xsl:value-of select="."/>
      <xsl:if test="position()!=last()">--</xsl:if>
    <xsl:for-each select="mods:name">
      <xsl:call-template name="name"/>

This was from MODS records that had mods:geographic elements, for instance, but no mods:topic,
mods:occupation, or mods:name elements.  
My solution was to wrap the dc:subject that is generated for mods:topic or mods:occupation elements, and
mods:name elements, in an xsl:test 
to make sure one of those exists before outputting the dc:subject element:

<xsl:if test="mods:topic | mods:occupation | mods:name">
    <xsl:for-each select="mods:topic | mods:occupation">
      <xsl:value-of select="."/>
      <xsl:if test="position()!=last()">--</xsl:if>
    <xsl:for-each select="mods:name">
      <xsl:call-template name="name"/>

I've attached a patch as well (which may more clearly illustrate the change).  If there is some place else (a
GitHub repository, for 
instance) to which I should report this, please let me know.

Attachment (MODS-to-DC-topic-fix.patch): text/x-patch, 1425 bytes