Guided CSV Importer + Validator via

Bulk Import Improvements initiative

Community Fundraising Effort: Improving the Bulk Import Experience for Hyrax and Hyku: Community-Funded Bulkrax Sprints

Prior to creating an importer and attempting the import process, Bulkrax (Bulk ingestion tool) should be providing feedback to the user in an easy-to-understand, actionable way. Further, there should be CSV templates based on the bulkrax configuration that the user can download.

Previously, the nature of the feedback was extremely technical, and difficult to understand, and was only generated once the importer is actually started.

All administrators

Slated for release and enabled by default in Hyku 7.2, but added to many production instances in Hyku 7.1.

Notch8

Presumably every single administrator, especially those new to Hyku

UofL

Amigos

PALNI / PALCI

Surfliner (UCSB/C/D)

Oregon State U.

Permalinks / Aliases / Redirects / Human-Readable URLs

Hyku previously required customization to support custom URL structures. It also required additional customization to support forwarding upon a changed human-readable URL.

This initiative has added the ability to support forwarding from previous URLs to canonical Hyku URLs, slugs, and added support for all this in bulk import. This is a win for both current users, and organizations migrating from previous systems to Hyku.

All tenants; especially institutions with large, long-lived collections.

This has been released to production, and included in Hyku 7.2

PALNI / PALCI / Pitt

UW

UTK

Amigos

UofL

Hierarchical / Nested Metadata

Frustrated by not being able to retain or define the order of Authors on a work? What about contributor vs creator, and nesting these metadata fields? One of the many problems we seek to solve here is to enable metadata librarians and administrators proper control and flexibility so that nested or hierarchical metadata fields can be displayed in a clear and consistent way to researchers and patrons, while providing proper attributions to actors associated with the respective work.

This is related to handling ORCiD profiles and that metadata schema as well!

Administrators, Metadata Librarians, Researchers

This is slotted for release in Hyku 7.2.

Many

Bulk Import Improvements (Bulkrax UX Improvements)

Community Fundraising Effort: Improving the Bulk Import Experience for Hyrax and Hyku: Community-Funded Bulkrax Sprints

Prior to creating an importer and attempting the import process, Bulkrax (Bulk ingestion tool) should be providing feedback to the user in an easy-to-understand, actionable way. Further, there should be CSV templates based on the bulkrax configuration that the user can download.

Currently, the nature of the feedback is extremely technical, and difficult to understand, and is only generated once the importer is actually started.

All administrators

Notch8

Presumably every single administrator, especially those new to Hyku

UofL

Amigos

PALNI / PALCI

Surfliner (UCSB/C/D)

Improved Work Creation and Editing UX

Creating single works can be a confusing and painful process for users who aren’t repository administrators. Many terms and relationships are shrouded in hyku-specific terminology. A researcher tasked with submitting their own works, especially if they are non-traditional researchers who may be submitting into custom work types, experience confusion and overwhelm.

Users unfamiliar with repository systems (e.g. researchers and students)

Repository administrators (in a support role)

University of Westminster

University of Westminster

Permalinks / Aliases / Redirects / Human-Readable URLs

Hyku currently requires customization to support custom URL structures. It also requires additional customization to support forwarding upon a changed human-readable URL.

The desired functionality, as we understand it, is that a given work or collection can have any number of supported URLs that may cover past versions as well as a primary (current) version to which all others are forwarded. It will also have a hyku permalink that would not be a human-readable structure, that contains a persistent UUID in the URL.

This would provide an easier and significantly more robust system for citation and migration.

All tenants; especially institutions with large, long-lived collections.

Repeated request across interviews; early adopters paid for customizations that did not re-enter core.

This would likely require some work to persist all associated URLs for works, and enable administrators to view / modify / delete them as appropriate.

UofL had this work done and is visible in our current live site, but it has yet to be brought into core code.

PALNI / PALCI / Pitt

UW

UTK

Amigos

UofL

Hierarchical / Nested Metadata

Frustrated by not being able to retain or define the order of Authors on a work? What about contributor vs creator, and nesting these metadata fields? One of the many problems we seek to solve here is to enable metadata librarians and administrators proper control and flexibility so that nested or hierarchical metadata fields can be displayed in a clear and consistent way to researchers and patrons, while providing proper attributions to actors associated with the respective work.

Administrators, Metadata Librarians, Researchers

This is related to handling ORCiD profiles and that metadata schema as well!

Ability to export all tenant configurations

• Migrations to the latest Hyku from much earlier versions / other platforms often difficult: enable easier migrations

• There’s a lot of styling overhead to starting from scratch with a brand-new tenant: enable new organizations to import a “jumping off point” - a fully-functional, mature tenant or organization’s configuration, including styling and custom CSS

• Testing the many different ways tenants can be configured is difficult, putting platform quality at risk: enable testing of tenant-specific behaviors / reproduce tenant-specific bugs

  • lay the groundwork for improved test automation with key testing tenant configurations + datasets (and known configurations)

Notch8 has started this work to augment the Hyku demo experience, as a sales enabler.

This is contingent on a migration from worktypes as customizable objects to a unified worktype with configurable sub-types. cc: @Rob Kaufman @Nic Don Stanton-Roark

Pitt

Notch8

Amigos

DOI Minting Workflow Enhancements

Streamlined minting (DataCite or Crossref), clearer admin UI, troubleshooting tools.

IR tenants; research libraries; any institution doing persistent ID management.

This will be part of the work undertaken as part of PRI / Enact with University of Westminster / CoSector

PALNI / PALCI

University of Westminster

Improved Support and Usability for Large File Imports (and Large Number of Files)

Currently, there is fear and uncertainty around large file (or quantity) uploads. The current functionality regarding chunked uploads often works. However, when it doesn’t, it is often difficult for users to understand what restrictions on uploads are from their admins vs which are technical errors or defects.

Repository administrators

Depositors

Libraries must be able to:

• import large files, one at a time and via Bulkrax. Currently, the system cannot import files over ??GB. Videos in particular can be this size and larger, as well as high-resolution images and maps.

• Import a large number of files via Bulkrax. Currently, the system has a difficult time importing over ?? files at one time. Libraries have had to chunk these files to get them into repositories. Technically, there should not be a limit to the number, assuming the infrastructure can support it.

• If there is a limit, it should be communicated on the upload page

UofL

Authority Controlled Vocabulary Improvements

Questioning Authority Gem improvement

It is generally a need to leverage multiple controlled vocabularies for a given metadata field – this is the way most (all?) metadata librarians often work. It’s difficult, time-consuming work, and represents an opportunity to make that work much easier.

• Enable Hyku to support mapping any metadata fields (e.g. Subject Heading, Person, Geographic Location, etc) to any number of authority sources (e.g. LOC, MESH, Wikidata, OCLC, Local) through a separate field on import.

  • Enable the Repository administrator to automate the mappings based on the metadata value itself (assuming it is unique), via the upload of a mapping file / or via a UI.

Metadata Librarians, Repository Administrators, Patrons and Researchers

DOT has some enhanced functionality in these areas, code needs to be extracted from their project and contributed back to Hyku main.

WVU

Presumably all users - this is just how it should work.

Amigos

Libraries already using ArchivesSpace will likely be looking for LCNAF integration

UofL

Pitt

IP-Based Permissions

Ability to grant or restrict access based on campus IP ranges; needed for reading-room models or licensed material.

Archives, special collections, and institutions with on-site access requirements.

Known gap; repeatedly requested for both IR and cultural heritage contexts.

UW

Tufts

PALNI / PALCI

Amigos

EZproxy Compatibility

improved functionality when proxied through EZproxy for off-campus users.

Institutions using EZ Proxy to enable off-campus access to materials with on-campus or institution-only access.

May be combined with IP-Based Permissions

UW

Willow SWORD API

Willow SWORD is a primary method for interoperability and ingestion of works and collections from external systems (e.g. Symplectic Elements, ProQuest). The user base and market interest for this API is increasing.

Right now, documentation, testing, and feature completion is relatively immature (we have MVP-level support for Willow SWORD API v2, and none for v3, and we need to continue testing and support for v2 resources with Flexible Metadata enabled.

We gotta continue the investment into the feature set and maintenance of it.

UW

Pitt / PALs

Custom Thumbnails

Right now, the default process for thumbnail generation yields suboptimal results (e.g. black screenshot for a video).

Submitting a custom thumbnail is preferable, and an ability of other repository systems (e.g. DSpace)

Amigos

Metadata-only should [gracefully] show for restricted (i.e. full-content) works.

In Hyrax, lots of it needs the metadata exposed to the public, but the content playback or images. Right now, it looks like an error:

https://library.osu.edu/dc/concern/generic_works/gq67k550p

Restricted public permissions ≠ no permissions. It’s like a different error handling.

All institutions with access-restricted collections

Discussion with the Samvera Partners on Dec 11, 2025

OSU

General Accessibility Work (Quarterly or Semiannual VPAT, plus remediation)

Regular review of accessibility compliance and structured plans for fixes.

All tenants; end users; administrators under accessibility mandates.

Spoken to across interviews. Some custom accessibility fixes have broken features; need consistent core updates.

Amigos

UofL

OAI-PMH Improvements & Efficiency Gains

More reliable harvesting, reduced failures, better performance for aggregators.

Add MODS

ETHOS, DPLA partners, institutions doing discovery-layer integration.

Needs stronger testing and documentation; frequently cited as fragile. H4C planning enhancements for user collection support; URL link back to work/file page.

Amigos

Pitt

Handle System Support

Support for persistent identifiers via Handle;

IR tenants; research libraries; any institution doing persistent ID management.

Raised by multiple institutions as missing infrastructure for long-term stability.

PALNI / PALCI

Amigos

ORCID Integration (read/write)

Link authors with their their research work; exchange data with the ORCID Registry to populate local author profiles; update ORCID records with publication information each time a repository deposit is made

Authors

ORCID has an API that can be used with repositories: Repository Systems - ORCID

Amigos

Custom header/footer

Allow repository administrators to customize the header and footer to include not just colors, but removing default links and adding their own. Allow each tenant to create their own custom header and/or footer in a multitenant environment.

Repository administrators

An Amigos library has asked for a custom footer in the multitenant environment.

Amigos

Restoring non-GA overlapping functionality of Reports section

With Google Analytics 4 functionality now functional in Hyku, the Reports section, which provided some different data, is blank except for the Users section. It would be nice to remove this section for now, and when decisions have been made on what data is needed to supplement GA4, bring it back.

Repository administrators, authors

A blank section in the Dashboard makes repository administrators wonder if there is something wrong with that section, or if there are others like in in other parts of the software. Here is a bit on what used to be there: Hyku Training: OverviewPreview

Amigos

Pitt?

Providing Link to IIIF Manifest

When preparing an exhibit of works, it would be nice to reuse the works' manifests in the exhibit, rather than downloading files and having to recreate them in order to embed them into an exhibit.

Curators

Possibly provide an icon or url to copy/paste or drag the link into an exhibit. (demonstration of this in action)

Currently reusing the manifest results in an error: Access to fetch at 'https://demo.hykuup-knapsack-staging.notch8.cloud/concern/scholarly_works/d7e4afa0-662c-4a81-92dd-ed8bd353066a/manifest.json' from origin 'https://uv-v3.netlify.app ' has been blocked by CORS policy: No 'Access-Control-Allow-Origin' header is present on the requested resource.

This is table stakes - JUST DO.

U of Washington

Item/page level metadata

Some parent/child relationships may have unique metadata at the child level. Hyrax’s NewspaperWorks gem accommodates this need, but it is not currently in Hyku core code.

Institutions with archival materials in Hyku

UofL had this work done and is visible in our current live site, but it has yet to be brought into core code.

UofL

Add text work type for textual materials that are not OERs or ETDs

Newspapers, journals, yearbooks, and the like are frequently found in archival collections. The current default work types and fields do not accommodate them (they have fields like volume and issue, as well as possibly using fields like abstract and table_of_contents currently only associated with oer and etd work types). Aside from other issues with the default URLs, it is not desirable to have a URL include the string generic_work, so adding additional fields to that is not optimal

Institutions with archival materials in Hyku

UofL

Improved sort options

Custom work in version 3 created additional sort fields, such as by source id. The default options of relevance, title, date uploaded, and date published do not accommodate ordering, say, newspapers by date, as sorting by title would put July before June alphabetically, and the other date fields are about back end functions. Adding this custom work back into core code would benefit any institution seeking to make browsing the front end more user-friendly.

Institutions with works that might best be accessed in a particular order

UofL had this work done and is visible in our current live site, but it has yet to be brought into core code.

UofL

Bulkrax Support for Fedora6 with Valkyrie

There are a lot of Bulkrax users on F6 / Hyrax. We want to broaden the userbase of Bulkrax as much as possible.

Fedora 6 users on Hyrax

Not a Hyku need, but a Bulkrax need, ultimately affecting the sustainability of Bulkrax as a dependency.

@Dolsy Smith I didn’t want to lose this need, so I put it here.

Enhanced UV media display

I would like to be able to attach multiple file types (still image, audio, video) to a single work and move between them in the viewer.

Repository administrators

PALNI/PALCI

Enhanced file manager options

I would like to add the ability to 1) exclude specific files from the files list but still have them display in the UV, and/or 2) disable download options for files in the list of files.

Repository administrators

This in some way needs to be interoperable with the download option built into the IIIF viewer. There is broad need to be able to disable downloading of specific files in a tenant, without configuring the entire tenant to be read-only.

PALNI/PALCI

UofL

Bulk updating of metadata in a specific field

Sometimes a repository or a linked resources changes its name or its URL. Sometimes LCSH updates a term and it will need to be changed in the Subject field across the repository. It would be convenient to be able to make these changes globally rather than collection by collection within a tenant.

Repository administrators

The subject term changes may be managed by improved authority linkage.

UofL

AI-assisted metadata generation

Create a free-standing metadata generation app 1) linked to a choice of LLM model, to 2) receive tenant-specific configuration data (CSV template, M3 yaml) for context, 3) read a set of files and 4) produce ready-to-import CSV that user then takes to their tenant.

Repository administrators; metadata librarians and archivists

PALs; Samvera AI WG