Library Answers: Los Rios Common Questions

This happens more often than we'd like. The reasons can vary, but one thing to understand is that Primo uses different processes to determine what shows up in search results as "Available Online" than it does to determine what full-text links to show (or not show).

Ex Libris distinguishes these two processes as discovery (allowing search results to show) and delivery (providing access via full-text links, physical holdings etc.).

Discovery is based on our Alma repository. A result will show because there is something in Alma--a collection, an electronic journal with specified coverage, etc.--that matches the record. Periodic indexing happens that results in these records being included in search results.

Delivery is based on a real-time check of what Alma calls services. When you go to the full record display or click "Available Online", Alma will take the metadata and match it against Alma data in order to provide one or more full-text links. In some cases this process fails.

Here are some reasons why discoverable records fail in delivery:

• Many of our records, including all article records, come from the Central Discovery Index (CDI). CDI is not updated every day. So if we remove a resource from Alma, it might continue to show in discovery for as long as 3 days, but if you click through to the full record, delivery will fail and the ILL form will show.
• For many of our EBSCO resources, we have a tool called the EBSCO Link Resolver plugin enabled. This tool does a background link test for EBSCO resources; if the test fails, the ILL form will show instead. This happens rather frequently for non-scholarly resources. Our thinking is that this is a better result for our users than sending them to an EBSCO error page, since if they send in the ILL form we should be able to get them the article fairly quickly.

In other cases we don't understand why delivery fails, and it is likely caused by some underlying issue that we can't solve locally. The ERM librarians frequently send these types of issues to Ex Libris support.

What can you do when you see this happening?

1. Check the source (e.g. journal) coverage by going to Journals by Title in OneSearch, and look for the resource in the source databases. Note that article titles may vary somewhat, since the source of Primo's metadata may not be the same as the database that holds the full text. Use the dates from the article record to browse the target database if appropriate.
2. Report the issue using the Report a problem button in the record. When you do this, the ERM librarians are notified and a permalink to the article is automatically included in your report, allowing them to reproduce the issue and either resolve it or send it on to Ex Libris support.

How Ex Libris explains this phenomenon

The following is copied from a presentation given on April 28, 2021, and is provided as an illustration of what we are dealing with here.

This can have several causes. The most common one is with metadata mismatches between CDI and Alma. In such cases the record is coming from an aggregator/selective package in Alma/SFX that is using the link resolver and the selective setting (We subscribe to only some titles in the collection) on the CDI Tab in the Electronic Collection Editor is set to No. This causes the real time availability check in Alma to fail if the metadata in CDI does not match the metadata in Alma, for example if there is no identifier and the title match fails as well. Sometimes this can be resolved by changing the metadata (our support team would have to do this) but you can also avoid this by using the selective option in the CDI tab and set this to Yes. Our content team is currently reviewing collections that are causing such issues to see if they can be treated as hybrid collections. For hybrid collections, if you do not use the We subscribe to only some titles... option, items that have no identifiers will use the link in record.