News

PBCore talks with the Open Metadata Registry

Recently, members of the PBCore Advisory Subcommittee decided that we wanted to know more about what the Open Metadata Registry (OMR) does, and how the folks involved with previous PBCore revisions intended to use it. On September 3rd, several subcommittee members called in and talked with Diane Hillman and Jon Phipps of the OMR.

After a lively round of introductions, Diane and Jon asked us to share the challenges of PBCore, about which they might be able to provide useful advice. We decided mainly to focus on charges of the PBCore Schema Team, like revising and enhancing the controlled vocabularies and eventually creating a Resource Description Framework (RDF) ontology for PBCore.

Diane and Jon suggested looking at more fleshed out controlled vocabularies for other standards that use the OMR, to see how the PBCore Advisory Subcommittee could maximize the potential of OMR. The OMR is an open system for creating and managing controlled vocabularies to support metadata interoperability via RDF and the semantic web. Members of the PBCore Schema Team who are focusing on controlled vocabularies will serve as administrators for PBCore’s OMR presence. The OMR is an RDF generator, which is intended for simpler models and designed for “the average user” to be able to generate a RDF ontology for element sets. It essentially provides an environment for generating and editing our vocabularies as a group.

Diane and Jon also gave us a tour of the RDA: Resource Description and Access Registry they’ve been working on, which provided an example of the additional work they do on top of the OMR. They created this site using GitHub and Git pages.

In terms of using the OMR to serve a particular standard, they recommend a workflow that takes the data from the OMR, automatically uploads it to GitHub, and then serves it out to another server, where the vocabulary is hosted.

They suggest that the best way to serve URIs is separate from the human readable documentation; for example, having the documentation for your users on your primary domain, and serving the URIs from a dedicated server or dedicated subdomain.

Overall it was a productive call and the PBCore team learned a lot that we can use in the future. We’ll be working on putting all of this new knowledge to good use, as we continue to improve PBCore. Thanks to Diane and Jon for taking the time to talk with us!

PBCore events at AMIA 2014

Are you headed to Savannah for AMIA this year? Want to get the scoop on current PBCore developments?PBCore-logoFinal

If so, go ahead and block off your schedule for Friday, October 10 from 11am – 1pm — it’s going to be a PBHardcore couple of hours!

First, members of the AMIA PBCore Advisory Subcommittee will lead a session titled “Pursuing PBCore: The Revitalization of a Schema and Community.” Casey Davis will introduce the session and speakers and generally introduce the current efforts of the recently established PBCore Advisory Subcommittee. Jessica Bitely will report on the results of the PBCore User and Non-user Survey, highlighting some of the suggestions from the user community as well as the misconceptions brought to light by non-user respondents. Jack Brighton and John Passmore will present on how PBCore is used at their organizations, and Mary Miller will discuss why her organization doesn’t use PBCore. Dave MacCarn will lead a Q & A.

Following the panel session, the PBCore Subcommittee will convene its Business Meeting from 12pm – 1pm. All conference attendees are welcome to join! Bring your brown-bag lunch (and your opinions!) and hear from members of each team discuss the specific efforts of the five established teams — Schema, Education, Website, Documentation, and Communication. The meeting will include 30 minutes for open discussion.

We look forward to seeing you in Savannah!

PBCore and Dance Heritage Coalition’s “Media Network”

Dance Heritage Coalition (DHC) is bravely and gracefully pioneering the field of dance preservation. As a consortium of institutions holding significant collections documenting the history of dance, they are the only national non-profit organization taking on the special challenges of this field. Those challenges include preserving a multi-format legacy and providing access to its riches while protecting copyright restrictions of the artists and choreographers. This is the aim of DHC’s online “Media Network”, a grant-funded project to create a union research database of dance-related moving images (searchable prototype available at http://archive.danceheritage.org). We were especially excited to learn that PBCore was being used to create their unique records, and grateful that Rebecca Fraimow, the hub manager for DHC’s New York preservation hub, offered to tell us how!

Rebecca and her fellow hub managers in Washington D.C. and San Francisco work with local artists to help them preserve their work and perform audiovisual conservation and preservation for the database at each of the hub’s digitization stations. So far they’ve made tremendous progress! While some member archives have performed digitization, the majority of film and analog tapes and born-digital files of one-of-a-kind materials, including performances and interviews, are sent directly to the hubs to be digitized and added to the database. The database currently holds 28,000 PBCore records mapped from MARC records and nearly 800 streaming video.

PBCore allows for each asset on the “Media Network” to include multiple, linked records, including the jpeg for the thumbnail image of the video still, the digital instantiation, for which technical metadata is captured automatically with the upload, the digital backup file, and a physical instantiation. PBCore data fields include form, identification, title, description, and relation to other material, which can be linked to contextual materials like posters, programs and reviews.

For description that is unique to performance, Rebecca and her fellow hub managers map the data to match PBCore fields where they can, adding their own attributes. For example, many of the records have long lists of creative collaborators, such as costume designer, composer, etc.

Streaming of the digital files is available through log-in afforded to member archives, libraries, and education centers in order to prevent unauthorized copying. However, the site makes discovery of the assets possible: the metadata on the front end is extensive and transparent, and anyone searching the union database – scholar, faculty, student, or the general public – can find the location of the physical instantiation.

PBCore might not seem like the obvious choice for the sole national dance-related repository, but in the absence of a performance-based schema, 2.0′s multi-part instantiations and breadth of fields for flexible mapping are supporting DHC’s most essential metadata requirements.

This fall Rebecca will bring her digital preservation skills and creativity to WGBH as an NDSR awarded resident and we are thrilled to welcome her! Congratulations, Rebecca, and thank you for sharing your PBCore experiences with us!

Written by Bryce Roe, intern for the American Archive of Public Broadcasting. Interview with Rebecca Fraimow, Preservation Hub Manager for Dance Heritage Coalition.

Continue to provide PBCore feedback through GitHub issues!

As the PBCore committee analyzes data coming from our survey (thank you, community!) we wanted to take a moment to let you know how we’ll be tracking community feedback going forward  – we welcome and encourage your continued input!

In a short few steps, you can sign up for a Github.com account and submit “issues” in the PBCore 2.0 repository. Github is a website used primarily by coders to track versions using a tool called Git, share open-source code, create documentation and build software. But if you’re not a coder, don’t let that intimidate you! In a few simple steps, you can help us use it to build a better schema (and perhaps spark an interest in using Github for other applications as well!).

To start out, point your browser to http://www.github.com/join

Next, choose a username and password. This should be individual to you, as Github distinguishes between individual accounts and organizational accounts.  After that, Github will ask you to choose a plan. If you’re a beginner user, we recommend the free plan:

Click continue, and Github will send a confirmation email to verify your account. Then, you’ll be all set to submit your PBCore issues!

To track issues, while logged in, navigate to: http://www.github.com/WGBH/PBCore2.0/issues

You’ll then be greeted with the interface below, where you can submit an issue for the PBCore schema committee to review. You can even follow discussion happening on the issue ticket as it happens!

The PBCore website committee will be assigning the Labels you see on the left, so need to worry about that.

Here is the interface when an issue is created. Please include a brief summary of your issue, and an extended summary below the header. You can take a look at existing examples for inspiration. The only hard and fast rule is specificity in articulating your need: we can’t address your concern without it!

If you would like to learn more about Git, Github and version control please comment on this post and we’ll follow up in a future blog with more information and resources.