Gabe,
Although I'd agree that dc.type isn't the best home for this
information, in simple DC and subsequently in oai_dc, it is a logical
solution and one that is being used. More logical than dc:description
in my view. But there is an issue in using metadata properties for
multiple things, e.g. type for 'resource type', type for 'status' etc.
This is why, in the Eprints Application Profile we created the status
property to support this [1]. As you have noted we also developed a
two-value vocabulary for PeerReviewed and NonPeerReviewed, also used by
the guidelines you cite at reference 4. This application profile is
relatively new and not yet widely used, but it attempts to address many
of the issues that using simple DC poses when describing more complex
resources.
Being consistent internally about how this metadata is captured is the
most important thing, along with controlling values with vocabularies.
That way you can expose metadata in different formats through simple
mappings.
Best,
Julie
[1]
http://www.ukoln.ac.uk/repositories/digirep/index/EPrints_Application_Profile#Status
Farrell,Gabriel wrote:
> I recently raised the question on oai-implementers[1] as to how items in
> our DSpace repository might best be labeled to distinguish those that
> are peer-reviewed. I received a variety of responses, including the
> following:
>
> * add a dc.description.peerreviewed field with value "Peer
> Reviewed"
> * add dc.description.PeerReview with values "is peer reviewed"
> or
> "is not peer reviewed"
> * dc.type="preprint" before review, dc.type="article" after
> review
> * the addition of "; peer-reviewed" to the value of the dc.type
> field
> * the incorporation of eprints.status="PeerReviewed" or
> eprints.status="NonPeerReviewed", as recommended in the
> Eprints
> Application Profile[2]
>
> I noted that I happened upon the UK dc.type qualifier vocabulary[3]. My
> institution could use that vocabulary for our dc.type values since all
> of our peer-reviewed items are already labeled as dc.type="Article". We
> could then reserve dc.type="Unrefereed Article" for the rare occasion
> when a published article is not peer-reviewed. Our main concern, after
> all, is the differentiation of peer-reviewed articles from theses and
> preprints. There are cases, however, where this strategy breaks down,
> such as preprints that have passed the peer-review stage. Also, I would
> want to make the peer-reviewed status more clearly evident.
>
> I believe the solution with the easiest implementation and best
> cross-repository compatibility is the addition of a second dc.type field
> with values "PeerReviewed" or "NonPeerReviewed", as suggested by Powell,
> Day, and Cliff[4]. A search on oaister.org for "peerreviewed"
> reinforces this conclusion. 72,851 records with Resource Type:
> PeerReviewed are returned (39,309 for "nonpeerreviewed"), suggesting a
> de facto standard.
>
> Any comments would be greatly appreciated.
>
> Gabe
>
> [1]
> http://www.openarchives.org/pipermail/oai-implementers/2007-June/001727.
> html
> [2]
> http://www.ukoln.ac.uk/repositories/digirep/index/Eprints_Status_Vocabul
> ary_Encoding_Scheme
> [3]
> http://library.princeton.edu/departments/tsd/katmandu/html/dctype.html
> [4] http://eprints-uk.rdn.ac.uk/project/docs/simpledc-guidelines/
>
>
> --
> Gabriel Farrell
> Library Systems Developer
> Hagerty Library
> Drexel University
> [log in to unmask]
> +1 215 895 1871
--
Julie Allinson [log in to unmask]
Repositories Research Officer
UKOLN, University of Bath, Bath BA2 7AY, United Kingdom
tel: ++44 (0) 114 2486457, ++44 (0) 1225 386580
skype: j.allinson
http://www.ukoln.ac.uk/repositories/
--
|