Metadata Call 2022-09-27

Time: 2:00pm-3:00pm Eastern

Call-In Info: Join our Cloud HD Video Meeting

Community Notes: https://docs.google.com/document/d/1WxZsUg2k4ItYqdQC1u-wiMrpNgp2rJ8gnIoHt4TbM2k/edit?usp=sharing

Moderator(s): Anna G.

Notetaker: Anna G.

Attendees: 

• Anna Goslen (UNC-Chapel Hill)

• Annamarie Klose (Ohio State University)

• Meg Wacha (City University of New York)

• Heather Greer Klein (Samvera)

• Amanda Hurford (PALNI)

• Julie Hardesty (Indiana University)

Agenda:

• Roadmaps Alignment Group Update (Annamarie)

  • Welcomed Rebekah Kati, new PO for Hyrax

  • Heather has facilitated this group, another may be stepping in

  • Proposal for core component changes. As we move toward Valkyrization, consideration of maintaining all gems. Will they be maintained across the board or are there some that individuals will have to choose to maintain locally?

  • Samvera state of technology white paper to be updated

  • Have a demo and beta of Valkyrized Hyrax by Connect, involves getting Hyrax 4.0 out of beta

• New IG facilitator starting at next meeting: Annamarie Klose (The Ohio State University)

• Meg: CUNY anticipating move away from Digital Commons, to Hyrax/Hyku. Would like a better sense of how metadata works on the back end. Uses simple DC, works in series. Wikidata - can describe works and people who authored works. Have faculty working at multiple institutions, or could be a faculty member working in multiple departments. Works can be in multiple collections in Hyrax

Discussion: Accessibility work in the metadata context

• What constitutes a “good faith effort” towards accessibility benchmarks?

  • Earlier in 2022 DLF group had a presentation about alt text. But discussion included broader topics. A lot of folks struggling with these topics. Concern about summarizing work in ways that may not be accurate

  • DLF doc: https://docs.google.com/document/d/1cRtizym4PxfDWfuHv3MlJPmSdSyhBrd2prNXh4EF5Gc/edit

• What are some efforts or projects in the wider metadata community that Samvera could explore or implement in this area?

• How can or should metadata be used in our applications towards addressing these issues? 

• How do digital repositories serve alt text? 

• How can Samvera applications serve alt text?

  • Annamarie: Have talked about using abstract to generate alt text, but question about the appropriate length. Same text could also come up twice if included as description and as alt text

  • Amanda: Created new worktype for OER - added a couple fields around accessibility. One summary field, one for hazards such as flashing lights. 

  • Meg: In Digital Commons, when faculty are providing descriptive elements, find they are willing to make the effort when related to discovery and probably also for accessibility purposes. 

  • Julie: Cultural heritage materials - metadata that comes with it can be very different from what you might need for accessibility

• Prioritization of metadata-related accessibility work in Hyrax

• Schema.org Accessibility Properties for Content Discoverability vocabulary

  • -primarily aimed towards EPUBs

  • -metadata for discovering the accessible properties of resource on the Web

  • -mostly machine-readable, expect for an “accessibility summary” property  that describes the accessibility features present in the publication as well as any shortcomings

  • -W3C community group: https://www.w3.org/community/a11y-discov-vocab/   

  • -final community group report (June 2022): https://www.w3.org/2021/a11y-discov-vocab/latest/   

  • -UX Guide for displaying Accessibility Metadata (September 2021): User Experience Guide for Displaying Accessibility Metadata 1.0   

  • -more information via DAISY Consortium about how to implement: Schema.org Accessibility Metadata   

• *Alt text and IIIF: A thorny and unfortunately not straightforward subject

  • has anyone successfully used the HTML ‘alt’ attribute to stuff alt text and/or OCR text content for materials being served by the IIIF viewer, as discussed in this thread: https://groups.google.com/g/iiif-discuss/c/XiQHaKpcSME?pli=1  

• Julie: one of first things addressed with IIIF is the controls and navigation

• AV with transcripts? New version of IIIF spec (3.0) would include this

• Annamarie: Describing different types of works can be a challenge. Abstract art, multiple photos of the same thing. Investment of time to create these descriptions

• Meg: Some work on using AI with images in Wikimedia Commons

• Annamarie- AI tools for AV? Julie- have tools at IU - MGM (metadata generating mechanisms)

  • AMP:https://www.nypl.org/digital-research/projects/amp-audiovisual-metadata-platform ; http://go.iu.edu/amp  

  • Makes use of Kaldi tool: GitHub - dericed/american-archive-kaldi: This repo houses open-source models for Kaldi speech-to-text software that have been trained on public media content.  

  • Study on transcription of AV from ARSC: Transcription of Archival Audiovisual Content: A Study of Accuracy and Cost

Next meeting: November 22nd, 2022, 2-3pm EST