Print

Print


On Friday, August 15, 2003, at 08:45  AM, [log in to unmask]
wrote:

> We need a XML schema for a metadata element set with certain functional
> requirements. The XML encoding should mirror our (planned) database
> structure as well as being compatible whith OAI-PMH (such that we can
> export our journal article metadata through that method).
>
> The phase I DOAJ service contains info on open access scientific
> _journals_ -- the phase II service should carry data on the articles in
> those journals.
>
> We would like to see how people encode article metadata in MODS, such
> that we  don't invent the wheel...

OK, I see.

The first point is that the current version of MODS doesn't allow for
detailed coding of articles.  This will change "real soon" (right
Rebecca?), however.  But it does mean that no one has really used these
facilities, aside in a testing context.  But the basic idea would be
that you'd have the article represented as a record, and the journal
itself would be coded in <relatedItem type="host">.  Here you'd put
volume, issue and page number info, coded something like:

<relatedItem type="host">
        <titleInfo>
                <title>A Journal Name</title>
        </titleInfo>
        <originInfo>
                <dateIssued>2003</dateIssued>
        </originInfo>
        <part>
                <detail type="volume"><number>23</number></detail>
                <detail type="issue"><number>3</number></detail>
                <extent type="page">
                        <start>34</start>
                        <end>54</end>
                </extent>
        </part>
<relatedItem>

If I get a chance, I'll send some sample files I've been working on, or
maybe post them somewhere.

I'm quite interested in this issue myself, as I am trying to convince
various free software projects to standardize on MODS.  To wit, it
looks like the bibliographic module for Open Office will -- at least in
the long term -- be upgraded to handle MODS.  Also, RefDB will be
moving to a MODS-compliant data model in the future.

What is your "planned database structure," BTW?  I wonder if there
could some common effort there?

Bruce