Publish Dublin Core records for digital inventory for Trove (National Library of Australia) using the OAI-PMH daily update

Set up OAI integration

1. If you already have this set up, you may only need to edit the IP range information to allow the NLA to harvest your records.
   Choose whether the OAI feed will be open to all or restricted by IP range. This setting applies to all sets published via the OAI.

2. In Alma configuration > General > External systems > Integration profiles Add Integration profile > OAI definition

• To restrict by IP range select Allowed IPs link in the Actions tab
• If you wish to have the feed open to all leave the Allowed IPs code table blank.

Create a set

1. Choose Record Format = DC
2. Choose any other fields/conditions to define the set for your institution
3. Optionally you may choose to use the Tag Sync National Catalog - this will depend on your cataloguing processes, and if you want to emulate the same process used for the Libraries Australia publishing profile.
4. Consider using a naming convention for your sets as they can be hard to find if you have a lot at your institution.
5. Make it a Public/Active set
6. Consider adding a “DO NOT DELETE without consulting xxxx” message in the description to avoid disappointment.

Create a publishing profile

1. In Alma, Resources > Publishing Profiles > Add Profile (General Profile) 
2. Naming (choose something meaningful for your institution that distinguishes it from other sets in Alma) eg. LA-DublinCore-digital
3. In the content section select the set you created
4. Output format = Dublin Core (Simple)
5. Publish on Bibliographic level
6. In the publishing profile section, select OAI
7. Consider a naming convention for your Set Spec and Set Name.
8. The set name is published as a description in the feed, so you may wish to include your institution name in there.
9. The data enrichment settings are quite simple.
   • Select "Add Digital Representation Information"
   • Specify dc:identifier in Repeatable Field
   • Select "Add Collection Information"

Managing your profile

1. Do not choose the Republish option in the profile without consultation with Trove Collaborative Services. This may create duplicate records.
2. You may choose to subscribe to success/fail notifications from Alma when the publishing profile runs
3. Specify an email address to Trove Collaborative Services for emailed reports when the harvest runs.

Contact Trove Collaborative Services

1. See useful links below for contact details
2. You will need to specify to TCS the thumbnail delivery URL syntax for your institution as the thumbnail URL is not included in the OAI-PMH output. TCS can use this information to collect the thumbnail URLs as part of their harvesting for display in Trove. It will look like: https://{ALMA_DOMAIN}/view/delivery/thumbnail/{INST_CODE}/{OBJECT_ID}
3. Advise them of an email address to receive notifications following each harvest

Withdrawing/Suppressing records

This is probably uncommon for digital resources which are unique to your institution, but it may happen from time to time. In the OAI-PMH output you will see the record status show as “deleted” if the bib record is deleted and when a DC record is suppressed:

<record>
<header status="deleted">
<identifier>oai:alma.INST_CODE:99207940902381</identifier>
<datestamp>2020-09-21T02:33:39Z</datestamp>
<setSpec>OA</setSpec>
</header>
</record>

Useful Resources

1. About the Data Contribution Services (OAI-PMH Daily Update Pilot) - National Library of Australia
2. Take 20 with Trove - OAI PMH Daily Update Pilot - National Library of Australia
3. Setting up OAI Integration - Ex Libris
4. Managing Search Queries and Sets - Ex Libris
5. Publishing and Inventory Enrichment (General Publishing) - Ex Libris
6. Thumbnail Delivery Service -- Ex Libris Developers Network

Caveats: At time of writing Trove Collaborative Services use of OAI-PMH is in a pilot stage and as such you can expect that not all workflows and requirements have been fully tested. There is some ongoing discussion in relation to handling deletions and updates of records for unique digital resources in the Australian National Bibliographic Database andTrove.

How to do this with MARC records