Recommendations On Herdc

Special Teleconference Meeting on HERDC issues for MACAR : notes and recommendations 8 May 2008

1. Attendance:

Simon Porter, Tom Ruthven, Katie Blake, Angela Lang, Kate Sergeant, Paula Callan (for Ann Huthwaite) Neil Godfrey, Joan Gray (Chair and note taker)

Joan welcomed Simon Porter, Information Manager (Research), eScholarship Research Centre, University of Melbourne to MACAR and Paula Callan, Repository Manager, Queensland University of Technology.

Joan also thanked everyone for attending at short notice.

2. Apologies:

Kerry Blinco, Jenny Millea, Helen Galatis, Alison Dellit, Suzanne Moir

Joan acknowledged the valuable comments received from Kerry Blinco and Helen Galatis.

Scott Yeadon has asked to be removed from the MACAR lists.

3. Review of HERDC documents at http://macar.wikidot.com

Joan noted that all HERDC/ MACAR documents are available at the MACAR wiki site http://macar.wikidot.com. Also all comments are posted on this site.

4. HERDC questions

Joan noted that it was important to be aware of international metadata standards and developments in our discussions and make recommendations as appropriate. Descriptive metadata describes the object/resource and agent or people metadata identifies and describes the agents associated with the objects/resources.

The recommendations are independent of any particular repository software solution. There may be in some issues optional recommendations/suggestions. It is up to each repository to decide how to implement these.

4.1.DEST category code

It was agreed that this code could be recorded and stored in a resource type or genre property and values could be assigned from a DEST category code vocabulary. Category code definitions should also be provided eg A1 – Book ; C1 – Journal article B1 - Book chapter etc

  • MACAR recommendation: Use type/genre property/field. Best practice is to use a controlled vocabulary.
  • MARC : 655 field – Index Term-Genre/Form subfield a and subfield 2
  • Dublin Core: dcterms:type or dc:type in simple DC
  • MACAR : macar:type

If the repository software cannot manage more than one resource type, or if the repository prefers not to make this data publicly available, a standard mapping from a resource type vocabulary used to categorise the resource (eg MACAR Resource Type vocabulary) to DEST category codes can be created to identify and collect this data for DEST/DEEWR purposes.

Paula Callan noted that a status value as in eprints.org would be useful. Peer review is relevant for journal articles and conference publications counted for DEST/DEEWR returns. A status vocabulary encoding scheme as in the Scholarly Works Application Profile could be used. This is a list of terms to indicate the peer-reviewed status of a publication. It is a simple vocabulary with just 2 terms – peer-reviewed and non-peer-reviewed and their definitions.

MACAR recommendation : Use type/genre property/field. Best practice is to use a controlled vocabulary.

  • MARC : 655 - Index Term – Genre/Form field subfield a and subfield 2
  • Dublin Core : dcterms:type or dc:type in Simple DC
  • MACAR : macar:type

4.2. Total number of authors and ranking of these

It was agreed that the total number of authors can be derived automatically from the resource descriptions themselves but author ranking may need to be recorded manually from the display in the record. However the display in the record may be misleading as some repository submission software may not allow the authors to be added and displayed in the order appearing in the publication. Some submission tools display the author inputting the data first. A data element indicating the preferred citation of the described resource may be required for HERDC as well as to support the further referencing and citing of research publications.

MACAR recommendation : Add a bibliographic citation field to the record
* MARC : 524 field – Preferred citation of described materials note – subfield a
* Dublin Core: dcterms: bibliographicCitation or dc:identifier in Simple DC
* MACAR : macar:bibliographic citation

4.3.Author affiliation

Repositories need to capture single authors and single/multiple affiliations and multiple authors and multiple affiliations in the metadata for research publications.

Joan indicated that most agent/person descriptions do provide an author affiliation attribute.

This relationship is usually expressed in authority data through links to authorized names and information notes.

Joan referred to the recent work of the IFLA Working Group (FRANAR) on Functional Requirements for Authority Data and the ongoing work of

of the DC Agent Working Group, and the FOAF specification (a Semantic Web initiative) These standards are in early development and we will continue

to monitor closely.

Another view is that the affiliation for a research publication eg journal article or conference paper is a property of the resource rather than that of the author. The author’s affiliation at the time the resource was created will persist even if the author moves to a different institution. If affiliation is then the property of the resource and not the author the resource description can contain multiple affiliations for multiple authors and there wouldn’t be a need to correlate particular authors and their affiliations.

Helen Gelatis suggested using DC author and publisher elements in the descriptive metadata and for multiple affiliations use of a agent schema like vcard ( See posting on the macar wiki site)

*

MACAR recommendation : Capture author affiliation in the descriptive metadata. Agent standards are in early development. Include in the descriptive metadata.
*

MARC: 110/710 - Corporate name
*

Dublin Core : Use dc:creator for author and dc:contributor for author affiliation (DCMI recommendation)
*

MACAR : Affiliated institution (university, faculty and school)

4.4.Total number of chapters in a book

There were a few suggestions for this item. It was suggested that DC description and qualifier table of contents could be used in repeatable fields. If using METS the division <div> elements within a structural map can be used to record the individual chapters of a book.

MACAR recommendation : To work collaboratively with the Research Office to determine where this data should be stored as it isn’t the kind of data that aids discovery of the resource but may only be required for administrative purposes and could be stored somewhere else.

  • MARC : 773 field – host item entry
  • Dublin Core: dc:description in Simple DC or dcterms: tableOfContents
  • MACAR : none

At this point there was a broader discussion of the various HERDC models being proposed and what metadata was suitable for storing in a repository and what was more appropriate for the Research Office to maintain.

Simon suggested that a web interface for researchers to maintain their own profiles should be developed.

It was also recommended that the data for HERDC be recorded once and both the repository and the Research Office work collaboratively to collect the outputs for HERDC reporting to DEEWR.

It was also noted that repositories wouldn’t make use of this data other than for this kind of reporting.

These recommendations only include MARC and DC mappings for institutions attempting to handle research management operations with their repository. It may not be advisable to try and ‘shoehorn’ purely HERDC metadata into a repository object. MACAR questions as to whether or not repositories need to record and store this kind of very specific purpose-driven metadata.

HERDC container or separate the metadata across metadata sections

Alternatively it was suggested that a separate HERDC container eg datastream in VITAL repository software could be used to keep all the HERDC metadata together Some institutions are already creating separate datastreams to capture and store this kind of data. The METS standard allows the packaging of digital objects with all kinds of metadata in a single file and this standard could be used to implement this. But finding a logical place for this in the administrative section of a METS file may be problematic. Also namespaces are a very important part of the METS standard and any XML schema incorporated in a METS package requires a namespace declaration for validation purposes. Further investigation of the METS standard is required to test if extension schemas such as this can be incorporated into a METS package and ingested into a repository.

4.5. Total number of authors within a department or school

Tom informed the meeting that this data was no longer a HERDC requirement.

5. Joan agreed to write up the recommendations and distribute to the group.

6. Other business : As there was no other business the meeting closed at 3.00 pm

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License