Metalogger

July 2, 2007

Electronic Theses and Dissertation Metadata Schema (ETD-MS) for Australia?

Filed under: E-Theses and ETD conference — Neil Godfrey @ 5:20 am

The NDLTD may not have the status of an international standard metadata schema for online theses and dissertations, but it certainly has established a lead throughout the US, Canada, UK, France and Germany at least. It is far from confined to the US. It is used by/used as a mappable set of elements with Library and Archives Canada (LAC), the EThOS’s UKETD-DC, Theses Electroniques Francaises (TEF), Germany’s XMetaDiss, and also within the Texas Digital Library Consortium’s MODS schema for theses. And much of the USA of course.

It certainly has a widely recognized lead. Nothing wrong with using something like ETD-MS in any Australian repository for starters. But are there any strong arguments against getting started with it or an application of it now? There is still a long way to go, since it is itself a minimal metadata set for basic ETD requirements. But it would be a strong starting point for getting Australia’s repository theses and dissertations potentially equally ranking in the harvesting stakes.

ETD-MS kicks in where DC leaves off, by applying the following metadata elements in addition to DC elements:

thesis.degree.name

thesis.degree.level

thesis.degree.discipline

thesis.degree.grantor

UKETD-DC uses the terms type.qualificationname and type.qualificationlevel to match the first 2 ETD-MS terms.

A MODS application employs these terms:

<mods : extension>

<etd:degree>

<etd:name>

<etd:level>

<etd:discipline>

and

<mods : name type=”corporate”>

<mods : namePart>

<mods : namePart>

<mods : role>

<mods : roleTerm>Degree grantor

The DC elements that these supplement in ETD-MS are:

dc.title
dc.title.alternative
dc.title.translated
dc.creator
dc.subject
dc.description.abstract
dc.description.note
dc.description.release
dc.publisher
dc.contributor
dc.contributor.role

dc.date
dc.type
dc.format
dc.identifier
dc.language
dc.coverage
dc.rights

In the above, dc.publisher is reserved for a commercial publisher in those cases where the thesis is actually (commercially) published. UKETD-DC uses the term “publisher.institution” for the thesis granting institution.

ETD-MS uses dc.contributor and dc.contributor.role for a thesis advisor or supervisor. Eprints and DSpace repositories also use the terms Supervisor (Eprints) and contributor.advisor (DSpace) for the supervisor/advisor.

There are still unresolved issues: e.g. the different natures of a particular thesis type — e.g. professional or coursework doctorates as opposed to research doctorates.

There still needs to be an ADT set for OAI harvesting.

And preservation metadata also needs addressing — a future post for that one.

MARC can be used for these ETD-MS terms or matching ones, too:

502 a for the thesis.degree.name

502 a for the thesis.degree.level (some tinkering will need to be done by IT staff to break these two elements, and other data below, into something sensible)

710 b for the thesis.degree.discipline

502 a or 710 a, b for thesis.degree.grantor

etd schema compared (corrected) — pdf file (you may have to refresh your browser screen to remove the old file display from the cache)

Advertisements

2 Comments

  1. I’m doing some information gathering and familiarizing myself with etd-repository issues; we’re in initial planning stages to do local theses using Dspace (apparently). We’ve also done some experimenting converting some locally assigned descriptive metadata to spreadsheets, then to DC/XML/marc. I was just looking at your etd schema (etd-md.pdf). I notice that for subject->dc:subject->653$a->to the MODS element, it has
    authority=LCSH. Is that intentional?
    From MODS to marc mappings v.3.2:
    653 $a with no authority attribute

    653 is uncontrolled subject term, not constructed according to thesaurus or controlled vocab rules; but I haven’t had any experience with MODS yet, so I’m trying to understand where lcsh comes in in the comparison.

    Thanks.
    Mia Massicotte

    Comment by Mia — July 5, 2007 @ 2:36 pm

  2. No it wasn’t intentional. Thanks for pointing it out. I created the chart by copy and pasting and deleting from a larger file, and should have deleted that lcsh reference of course. 653 is for uncontrolled keywords, as you point out. Thanks for alerting me to this. Will correct it.

    Comment by neilgodfrey — July 5, 2007 @ 10:34 pm


RSS feed for comments on this post.

%d bloggers like this: