UC Library Search (Primo VE) Staff FAQ

Previously known as the UC Library Search Information Page.

How to use this guide:

This page is intended to be used as a reference guide. If you have any questions or suggestions for new content, please contact helpbox-library@berkeley.edu.

Known issues in UC Library Search

• Issue:  None of the article-level links for content in Factiva will connect to the articles; they connect to an error page.  SCP is working on resolving.
  • Workaround:  go to Factiva from the A-Z databases list
  • Estimated fix:  TBD
• Issue: Some journals are not appearing in the Journals Search feature.
  • Workaround: Use the Articles, books, an0d more search instead
  • Estimated fix: TBD
• Issue: Many eresources we should have access to are not available in UC Library Search
  • Estimated fix: TBD. CDL has not been able to activate many Tier 1 and 2 ebooks and electronic resources. A project team composed of members of CDL as well as UC campuses has been tasked with addressing this high-impact, widespread issue experienced by all ten UCs.
• Issue: Approximately 800 database-level records migrated in to Alma as suppressed at the advice of Ex Libris
  • Estimated fix: TBD. The CDL Acquisitions team is prioritizing this cleanup. As new collections with portfolios are acquired and activated, all will be unsuppressed by default.
• Issue: CrossRef is experiencing system problems
  • Estimated fix: TBD. This impacts the DOI lookup for PrimoVE/Alma. OpenURL pages that rely on this lookup (such as those from Google Scholar) may appear with incomplete citation information while the DOI lookup is not functional.
• Issue: Linking from FirstSearch not always landing on UC Library Search record
  • Estimated fix: TBD. Due to differences in OCLC number formatting in bibliographic records, our settings in OCLC First Search may not always successfully search a known item in the UC Library Search Catalog. Systems and Discovery is investigating the issue.
• Issue: Some records are erroneously marked as Open Access when they are not
  • Estimated fix: TBD. Please see Why do some Open Access records lack URLs?
• Issue: Searching by call number does not always work as expected
  • Estimated fix: TBD. This could be related to a number of factors, including a difference in the item and holdings call number, or how call numbers are indexed in Primo VE. Further information on how call number searching works is included in the question How do I search or browse by call number?
• Issue: Records for completely different records are merging in UC Library Search display
  • Estimated fix: TBD. Individual instances of this occurring can be reported to helpbox-library@berkeley.edu to pursue resolution. Further information on this behavior is described in Why are some records and holdings unexpectedly merging in UC Library Search?
• Issue: Screen readers are not working optimally in all browsers.
  • Estimated fix: TBD. We have reported this issue to ExLibris, and shared it with other campuses. ExLibris has confirmed the issue, but we do not have a forecasted resolution date. If you notice any accessibility issues in UC Library Search, please contact helpbox-library@berkeley.edu.
• Issue: Searching some non-Roman scripts do not function as expected in UC Library Search advanced search
  • Estimated fix: TBD. The Discovery Working Group is working together with several librarians with expertise in non-Roman languages to test and create an inventory of all the related issues for searching in UC Library Search.
  • Update: The “Title starts with” search in advanced search does not query the MARC 880 fields, which contain non-roman scripts. Using this field will not effectively search these fields.

General information

What are our accessibility standards?

Primo VE (UC Library Search) attempts to meet Level AA of the W3C Web Content Accessibility Guidelines (WCAG 2.1) and Section 508 of the US Rehabilitation Act for features and functions. However, there are many gaps in accessibility in Primo VE and we are advocating for development with ExLibris. See the ExLibris documentation for further information. If you notice an issue related to accessibility please submit feedback to us via helpbox-library@berkeley.edu.

Who do I contact if I need help?

If you need help using UC Library Search, or have a question, please use the Alma/UC Library Search Problem Reporting Pathways for Library Staff to determine next steps. 

Why is the response time so slow?

This is a known issue in UC Library Search. ExLibris is looking into the issue.

Will the system ever be unavailable?

Yes. Ex Libris has a standard maintenance window (Saturday 11:00AM - 9:00PM), though typically they use a much shorter window for maintenance and release upgrades. Releases and release updates occur each month. If there is a longer than forecasted or off-cadence system downtime, we will communicate it with staff. More information can be found in the Ex Libris Reliability documentation.

How can I make suggestions for enhancement requests?

The Issues + Ideas Tracker empowers staff to share non-urgent issues and ideas for UC Library Search such as enhancement suggestions, frustrations, comments, or questions —  in a low barrier format. This tool is designed for you to capture issues and ideas that impact the user experience. APriCoT reviews submissions regularly and communicates updates in SILS dispatches.

The Discovery Working Group (DWG) reviews, recommends, and tests updates to the UC Library Search User Experience (UX) in consultation with APriCoT and other  SILS groups.

ExLibris pushes monthly updates to both Primo VE, the software underneath UC Library Search, and Alma. This provides an environment of continuous improvement and change, including bug fixes and deployment of enhancement requests.

Can users still search WorldCat?

All UC campuses have access to UC Worldcat (also known as Basic WorldCat Discovery), a product that is linked to UCs’ FirstSearch subscription. Each individual campus also has their own campus instance; at Berkeley, it’s called Berkeley Worldcat. URLs from Melvyl will redirect to UC Worldcat. Though these two products are available, we encourage users to start with UC Library Search for their searching needs. 

There are two different ways for UC Berkeley users to search Worldcat.

1. In Advanced Search, users can select the Worldcat radio button to use the Worldcat search profile. This search uses an API that retrieves records from Worldcat.org. The results will not include any facets or filters. If users would like to work with their results using facets or filters, they will need to click on the Worldcat button on the left side of the results screen to continue their search in the Berkeley Worldcat interface.
2. Users can go directly to https://berkeley.on.worldcat.org. From this interface, users can use the “Get it at UC” button to locate materials at the UC Campuses as well as make interlibrary loan requests for items not owned by Berkeley.

Are UC Library Search URLs permalinks?

The URLs created in the full record view are “direct links,” not “permalinks;” we cannot guarantee that these URLs will remain unchanged in UC Library Search due to merged records, record removal, or other updates which may affect the URL. In July 2022, the verbiage in UC Library Search has been updated from “permalink” to “direct link” for clarity. Please note that these direct links still use the word “permalink” in the URL string.

If you want to link directly to records and are willing to maintain the permanence of those URLs, you could get a short URL maker account and maintain the URLs yourself. This way you don’t have to update the libguide and you update the URLs in shorten URL tool. If you would like this, please submit a ticket to helpbox-library@berkeley.edu.

Record display information

What does “may be available” mean? 

The “may be available” status is assigned to holdings records which do not have any items attached. Multiple holdings were created for some records as a result of the data migration from Millennium into Alma since holdings are handled differently between these two systems. This is a known issue, and has been identified as a post-migration clean-up task.

What does the “technical migration” or “technical” status mean?

The “technical migration” status was applied to all items which had a status in Millennium which did not directly map to a status in Alma. This status clears when an item is checked in, and we anticipate that most of the items with the “technical” migration status will clear after go-live.

What does “out of the library” mean?

“Out of the library” is the term for items which are not currently on the shelf.

What are “Host Bibliographic records”?

Millennium had the capability to link an item to multiple bibliographic records. This was used both for bound-with and analyzed sets. Alma does not support this in the same way and instead creates a Host Bibliographic record with the Related Titles (MARC 773) field in the full view of bibliographic records. UCB groups will be looking into this post go-live as there is nothing that can be done during this stage of the project. 

Why am I not seeing the expected resource type?

This is an outcome of the data migration for Test. The resource type is mapped from a constellation of bytes and MARC fields in the bibliographic record. This is a known issue, and we plan to update material types for records after go-live.

Why does the loan information (i.e. “loanable” or “not loanable”) sometimes change for an item when a user is signed in?

ExLibris created documentation which explains the following behavior: “When a user is not authenticated, Alma does not have all of the information required to accurately determine which Fulfillment Unit rule will be applied.  Often the Loan Terms of Use are sensitive to user information, such as User Group. Because of this, Alma has to assume a TOU might apply, thus marking the item as "Loanable."

Why do some Open Access records lack URLs?

We have identified this as a known issue in UC Library Search. Many of the Open Access records in UC Library Search are a part of the Central Discovery Index (CDI), which means that content providers outside of ExLibris provide access and metadata for these records. So, if an error or omission is present, it cannot be fixed locally by UC Berkeley Library staff. If you notice an issue, please submit a ticket to helpbox-library@berkeley.edu. Systems and Discovery Services staff will then open a ticket with ExLibris to request the content provider to provide an update to the record(s).

Why do some records lack item records?

You may notice some records in UC Library Search which have the title in call capitals, are “on order,” minimal bibliographic information, and no attached inventory. These records are likely Embedded Order Data (EOD) records, which we receive from vendors and load to Alma before the item is received or while it is being processed by staff. It is expected that these records are discoverable in UC Library Search. If you have questions about a particular record, please contact helpbox-library@berkeley.edu.

Can we search serials in a way that will get around scrolling through years and years of holdings?

Currently, using the description filter embedded in the GetIt display in UC Library Search is the most efficient way to search through a serial holdings. 

How can patrons make Alt. Media scanning requests?

A new pick up location “Print Disabilities Services (Doe Library)”  was recently added to the pickup location drop down menu. This is a change to make it easier to distinguish Alt Media scanning requests. These items will go to Digitization Services for scanning for patrons that are eligible for these services. Please see the Library webpage for more information on Print Disabilities Services.

Searching information

What’s included in UC Library Search?

In the default scope called “Articles, books, and more,” print, digital, and electronic records from all ten UC campuses, including Berkeley, are available to search along with a database of citations and linked resources called the Central Discovery Index (CDI). ExLibris provides a collections list of the resources included in the CDI, which is updated regularly.

The scope “UC Berkeley Catalog” contains only print and digital material belonging to Berkeley. This scope is similar to OskiCat, but contains fewer electronic records and is designed to surface known items in Berkeley collections with fewer clicks. A much larger universe of electronic resources than previously available in Oskicat is discoverable through CDI in the “Articles, books, and more” scope.

The “UC Berkeley Special Collections and Archives” scope contains only records from special collections and archives locations across the UC Berkeley campus. This scope also provides a shorter path with fewer clicks for novice and advanced users who want to find primary source or special collections materials on campus.

Why can’t I refine my results to specific campus libraries? 

There is a known and reported issue related to the intermittent display of facets in UC Library Search. As a result, you may have trouble limiting your results to specific campus libraries. ExLibris has indicated that this should be fixed in the August update of the software.

How do I search or browse by call number?

Call number searching works differently in UC Library Search than it worked in OskiCat. Call numbers are keyword searchable in the main search scopes. When you use the Browse Search feature for either Library of Congress or other call number, you are searching for the call number in the holdings record. The holdings call number, or “permanent call number”,  is mapped from the item call number, and this is what we are used to searching in OskiCat. Through this method, you should retrieve a browse list of the available call numbers because each item has a call number assigned. Alternatively, if you use the advanced search option, that pulls from the call number in the bibliographic record, and not all bibliographic records contain a call number. So, the advanced search may not retrieve all call numbers if searched.

Additionally, due to an outcome of migration, there may be a difference between the “permanent call number” and  the item call number. This is especially true for special collections. Users can view the item call number in the full bibliography display in the GetIt section with the other item information including barcode and location, but the item call number cannot be searched in Primo VE.

Why am I seeing so many versions of the same work in the search results?

We suppressed FRBR in our search results. FRBR (functional requirements of bibliographic records) groups records of the same expression together, much like results in Melvyl. Right now, we have suppressed FRBR for everything, but sometime after go-live, we will be selectively suppressing FRBR for specific collections where known issues exist around FRBR inaccuracy, including music and performing arts materials, law, and art books.

Why do I see results that don’t include my search terms?

Primo VE looks for the keywords entered in a search in a variety of indexed fields. In addition, it may perform additional processing of your terms that include stemming, term variations, normalization, and/or query expansion based on controlled vocabulary. As a result, many of the most relevant items that are returned in a search result set will include your search terms somewhere in the full record and you may also see results which do not include all of your search terms. 

When trying to figure out why a result is being returned, it can be useful to view the full source record, which may include the fields that are being searched even if they are not being publicly displayed. This can be done using the &showPnx=true URL extension. When postpended to the URL of a record, this extension reloads the page and shows you the full source record. To use the extension, navigate to a record in UC Library Search and in the URL bar, add &showPnx=true to the end of the URL (e.g. https://search.library.berkeley.edu/discovery/fulldisplay?docid=alma991051196469706532&showPnx=true)

Note that search in our own library records operates differently than search in CDI records; read more on Performing Basic Searches in Primo VE and Search and Ranking in the CDI. 

How are the results ranked?

UC Library Search uses several factors to determine the ranking of results in the UC Berkeley Catalog Scope: 

Relevance: Greater weight is given if keywords are found in author, title, or subject fields, and an exact title match is set to be weighted the most heavily. Frequency of search terms in the record is also considered, as well as the term or phrase order. 

Academic significance: Greater weight is given to items that are published in a peer-reviewed journal and how many times it has been cited.

Type of search: The system can interpret a search in a few different ways and ranks results accordingly. If it thinks the user is performing a broad topic search, overview material will be ranked higher in the results. If it thinks the user is performing a known-item search, author and exact title matches will be higher in the results. If it thinks the user is pasting an entire citation into the search box, it focuses on the author, title and date information to find the item.

Publication date: More recently published items are ranked higher.

Institutional boosting: Because we are operating in a consortial environment, we have also chosen to “boost” Berkeley’s items so they rank higher than items held at other campuses.

The relevancy ranking system works slightly differently in the Articles, books, and more profile. When a patron issues a search query in Primo backed by the Ex Libris Central Discovery Index (CDI), the query is issued to both the local index and CDI. Search results from each of these indexes are ranked according to their relevance ranking algorithms, and they are blended to form the final search results presented to the patron. This document discusses the relevance ranking algorithm used by CDI.

Relevance ranking in CDI is determined according to a continuously tuned, proprietary algorithm, and is built on a foundation of two building blocks: the Dynamic Rank and the Static Rank. The Dynamic Rank is a collection of relevance factors that represent how well a search query matches each record, and the Static Rank is a collection of relevance factors that represent the value or importance of each record. Both of these are important in determining the ranking, and top results need to have good scores from both Dynamic Rank and Static Rank. For more information on CDI searching and ranking, please view the Ex Libris document: CDI Search and Ranking.

Why do I find things at other campuses that are not available to me? 

Some records for items such as equipment (macbooks, for example) from UC campuses surface in the “Articles, books, and more” scope. These records are not actually eligible for borrowing through the Automated Fulfillment Network (AFN). Currently, SILS is actively working to determine a method to prevent these records from surfacing to prevent confusion.

How are records marked as Open Access?

Content in the Central Index is considered Open Access if it meets the following general criteria:

• An item is freely available and openly accessible without requiring authentication by the user.
• An item is identified by the provider/publisher as Open Access.
• An item resides in a known Open Access repository, database, or journal collection that we determine to be Open Access.

This definition, as well as further information on how Open Access records 

display and function in Primo VE, can be found in the ExLibris documentation.

What does “Held by library” limiter mean in the Availability facet?

Using this limiter will allow you to search print materials held by UC Berkeley and the other UC campuses in the “Articles, books, and more” profile. UC Berkeley print materials can also be found using this limiter in the “UC Berkeley Catalog,” and “UC Berkeley special collections and archives” profiles.

How does UC Library Search handle punctuation in queries?

Punctuation works differently in UC Library Search than it did in OskiCat. Straight quotes are recognized, but curly quotes are not. The asterisk works for multi-character truncation (i.e. cultur* = culture, cultural, culturally), and a “?” can be used as a single character wildcard (i.e. (wom?n = woman, women, womyn). Also, UC Library Search, Alma, and WorldCat the dash should be replaced by a space. For example, "year-book" has to be searched as "year book" and not "year-book" or "yearbook." It should be noted that dashes do not work in WorldCat queries; for example a search for “COVID-19” will yield no results, but a search for “COVID 19” will work. More information on search strategies can be found in the Cheat Sheet for Staff compiled by the User Outreach and Internal Training team.

Where did the Stanford, Center for Research Libraries, and HathiTrust buttons on the results page go?

In OskiCat, these buttons were used to assist searching in these collections. However, in 

UC Library Search, we are using a different searching paradigm. The aim of the “Articles, Books, and More” profile integrates more resources, both print and electronic, from UC Berkeley to the UC consortia and beyond. This profile will include CRL records specifically in the search as part of the Network Zone. We do not currently anticipate adding any further facets/buttons/profiles in UC Library Search.

How do I limit my results to resources only available online?

While performing a search, you can use the “Available online” filter on the left side of the screen. After checking the box for “Available online” under the “Show only” facet, click the green “Apply Filters” button to retrieve an accurate set. 

Why are some records and holdings unexpectedly merging in UC Library Search?

There are several reasons why you may see two records merging unexpectedly in the UC Library Search display. In some cases, the issue could be related to our current dedup configurations. First off, what is dedup? According to the ExLibris documentation, "the duplication detection or matching process is based on creating a dedup vector for every Alma record (and all imported records from external sources). The vector includes all the data required by the Duplication Detection algorithm to determine if two records are equivalent.  The vectors include one or more keys that identify the record. Keys are based on the metadata fields in the source record (MARC or DC records)." 

This means that you may see two separate Alma records grouped together in the same display if the dedup algorithm determines they are a match. However, it should be noted that these two records are not merged in Alma. We currently have our Primo VE dedup configuration set to electronic preferred record to align with the UC-wide Discovery Operations Subteam recommendation to update the preferred record to electronic. This may resolve some issues, but not all. If you notice records merging unexpectedly in the UC Library Search display, please submit a ticket to helpbox-library@berkeley.edu for investigation.

What are the differences between the options in advanced search?

ExLibris sketches out the differences between the options in advanced search in their documentation. 

• Contains exact phrase – Returns results that contain phrases that exactly match the phrases specified in the query. 
  • Results for title searches may include records that do not include the search term in the title. This is because exact Title searches are expanded to include the record's alternative title and authority subject fields.
• contains – Returns results that contain all words in the phrase, but the words may be in a different order and may not be as close together.
• starts with – Returns results that contain words that start with the specified string. This type performs left-anchored title searches only. When performing a Title search, it is recommended that you do not omit any leading articles from the title. For example, The Oxford Handbook is preferred to Oxford Handbook.