Kia ora Stuart!

You may be interested in a couple of OAI-PMH providers I wrote not that
long ago.

The code is here: https://github.com/Conal-Tuohy/Retailer and there are a
few posts about it on my blog http://conaltuohy.com/

Note that the providers are not available online publicly, but you can set
them up yourself without much effort. The data they provide is sourced
externally, via a web API, and in that sense the implementations are
certainly "different". They are stateless transforming web proxies,
translating between OAI-PMH and upstream custom Web APIs, implemented as
XSLT stylesheets hosted in a simple Java-based web server called Retailer.

The upstream data providers are the digital newspaper collections of the
National Libraries of New Zealand and Australia.

The content types supported are also unusual; as well as the mandatory
oai_dc, both the OAI-PMH providers provide records in XHTML format. These
records are in fact the full text of the newspaper articles, with basic
metadata embedded in the HTML head.

In addition, they also proxy the native data formats provided by the
upstream API ("trove" and "digitalnz"), with a few minor changes, such as
to add the XML namespaces required by the OAI-PMH spec.

A couple of years ago I wrote an OAI-PMH endpoint for a local instance of
the archaeological repository software tDAR. This is written in Java, using
the Struts framework. Here it is on the American tDAR site:
https://core.tdar.org/oai-pmh/oai?verb=ListMetadataFormats which provides
MODS.

Further to Bernadette's suggestion of RIF-CS (research metadata) from
Deakin, you can also reharvest this metadata, and metadata from other
providers, from the Australian federal agency ANDS. See
http://developers.ands.org.au/services/collections-registry-api/oai/

You probably already have this one: the EAC-CPF records available from the
National Library of Australia:
http://www.nla.gov.au/apps/peopleaustralia-oai/OAIHandler

Con

On 7 November 2014 08:53, Stuart A. Yeates <[log in to unmask]> wrote:

> I'm looking for a unusual OAI endpoints (different implementations,
> different metadata schemes or extensions to schemes, different
> structures, unusual content types, etc) to test against. I'm aware of
> the list a couple of mainstream lists of which
> http://www.base-search.net/about/en/about_sources_date_dn.php?menu=2
> is the most comprehensive  the and the live demos of dspace, eprints
> and fedora. But I'm looking for more obscure installs and corner
> cases.
>
> Does anyone know of any other candidates?
>
> Implementations known to be buggy, broken or dubious especially welcome :)
>
> I'll publish a list of endpoints I find useful.
>
> cheers
> stuart
>