Metalogger

November 20, 2007

More MODS advantages over MARC

Filed under: MARC,MODS — Neil Godfrey @ 1:34 pm

I have recently had opportunities to work with MODS in creating templates for research and publications data external to the university repository environment.

Multiple note fields can be added each with its own customized display label in MODS. One can add multiple 500 general note fields in MARC but MARC does not recognize a $i display text to differentiate them.

Example:

<note displayLabel=”Objectives” />
<note displayLabel=”Background” />
<note displayLabel=”Methodology” />
<note displayLabel=”Progress” />
<note displayLabel=”Implication” />

In the same way displayLabels can be used to distinguish quite different “title” types — e.g. the titles of research program and subprogram as opposed to the title of specific research activity.

And multiple affiliations and addresses can be nested with each personal name in MODS. MARC allows only one — NR (not repeatable) — $u subfield for an affiliation or address in each 100 or 700 field.

<name type=”personal”>
<namePart />
<affiliation>$researcher[‘affiliation’]</affiliation>
<affiliation>$researcher[‘address’]</affiliation>
</name>

Those are TWO HUGE advantages as anyone trying to squeeze nontraditional data into digital archives will appreciate.

Advertisements

2 Comments

  1. Your statement about squeezing non-traditional data into digital archives argues against having any traditional schema in a repository. I am thinking more and more along these lines. XML was meant to be liberating, and yet we are continuing to shoehorn data into schemas that were not designed for it. The issue of repeating property is a prime example of this. Shouldn’t every property be repeatable if you need it to be?

    The only reason you need to have a schema for descriptive metadata is where you are going to integrate it with another repository or database using that same schema.

    Comment by Katie Blake — November 20, 2007 @ 3:52 pm

  2. Sounds like you are arguing for RDF. Yes? Yup. (And an RDF base would in theory enable any schema to be created any time it was needed.)

    Comment by neilgodfrey — November 20, 2007 @ 3:57 pm


RSS feed for comments on this post.

%d bloggers like this: