Hydra Connect 2016: Presentations, panels and breakouts


Tuesday - Plenary track

Morning session in the Rabb Lecture Hall



Alberto Martinez (Daniel Cosío Villegas Library, El Colegio de México,Mexico City)

This presentation aims to present opportunities for collaboration between current Hydra members and The Daniel Cosío Villegas library at El Colegio de México (COLMEX) as well as other Mexican institutions that are interested in making a Mexican-based Hydra users group. COLMEX has been actively planning the implementation of a Hydra-based repository. Given that COLMEX has a significant presence in various digital libraries and repositories interest groups in Mexico, we have taken the opportunity promote Hydra as an alternative, not only with the hopes of promoting the project but to find local partners that might be interested in collaborating. We hope that we might find international partners who will help to spur initiatives through various means of evangelizing, helping support efforts, and perhaps coming down to Mexico to visit. In this manner we can help Hydra become a truly global initiative and one which considers north-south collaborations especially those outside the English-speaking world.


Progress report: Hydra in a Box (Video Link)

Hannah Frost, Mike Giarlo

An update on recent progress on the Hydra in a Box project, including work related to product development for the repository and metadata aggregation components, development of the hosted service, development and infrastructural decisions, and community engagement.

Slides: http://bit.ly/hybox-update-hc2016


State of the Hydrasphere (Video Link)

Tom Cramer, Debra Hanken Kurtz

This annual report on the Hydra Project will provide a synopsis of the project’s current state from a high level perspective, including recent developments and important trends in adoption and activiity, the technical framework, the community framework, major projects and milestones, and where we may be going in the near future. With so much activity in so many different parts of the project, this session is a chance to take a step back from the many trees to survey the whole forest of the HydraSphere.  

Slides: https://docs.google.com/presentation/d/1IZCOaYtrsdefdG75Z0fZ-g8ED2grzR82dNtDs0NupbM/edit?usp=sharing


Wednesday - parallel tracks

The full list of Lightning Talks is here.  All take place in the Rabb Lecture Hall.  

Please note: Each session listed below has a suggested "audience" but this is just for your general guidance in choosing between topics.  Anyone attending Hydra Connect is more than welcome to attend any of these sessions!

Approaches to asynchronous storage and Hydra

Breakout - Don Brower, Will Cowan, ANO

Audience: DevOps and managers

The Hydra stack does not expect content to reside on high-latency storage or to be temporarily unavailable. Fortunately Fedora and Hydra are flexible enough that workarounds are possible. This session is to discuss approaches to and experiences with handling tape storage and integrating these systems into Hydra repositories.


Archiving Research Data into Hydra through the Open Science Framework (OSF)

Presentation - Rick Johnson, Don Brower, Sayeed Choudry, Elliot Metsger

Audience: All

Archiving Research Data into Hydra through the Open Science Framework (OSF) - A look at initial work of Notre Dame and Johns Hopkins to archive research projects from the OSF into Fedora and Hydra repositories, and first implementation of a Fedora Research Object Model. This plugs into a service offering of the Center for for Open Science, OSF for Institutions (OSF4I). ND/JHU version will be initial support for OSF Fedora Archiving Add-on in the OSF as part of OSF4I offering. We hope to start discussion around next steps for other Hydra institutions to use this along with OSF4I to allow them support to archive research data from the OSF into their own Hydra/Fedora repository.

Slides: (Google)  JHU-ND HydraConnect 2016 slides (1).pdf , Audio: (MP3)


Avalon's road to an SAAS pilot

Breakout - Debs Cane, Phuong Dinh, David Schober, Jon Cameron

Audience: All

Avalon Media System is an open source software system, originally developed by the libraries at Indiana University and Northwestern University, for managing and providing online access to large collections of digital audio and video for purposes of teaching, learning and research. Avalon was initially released in 2013, and as of early 2016, there are at least six adopters of Avalon but no consistent pathway for adopters to contribute financial or development resources to the project.

In 2015, we were awarded a grant from the Mellon Foundation to create a long-term sustainability plan that includes better integration with the Hydra community, a strategy to increase adoption, and a pilot of a software-as-a-service (SaaS) offering in addition to its current locally-installed model.

This presentation will provide an outline of Avalon’s path forward and how it corresponds with directions of the Hydra and Fedora communities. We are radically changing our development strategy as we move forward, and we will discuss our strategic initiatives and outline new areas of exploration and development. Additionally, we will discuss the work in progress as we move forward with our SAAS pilot.


Avalon – Moving features to Gems

Breakout - Debs Cane, Maria Whitaker, Jon Cameron

Audience: not specified

Avalon currently has several features that, if they could be shared, might be of interest to other projects within the Hydra community. Our team is focused on moving some of these features to ruby gems to be contributed to the community for use in other hydra projects. But we’d like the Hydra community to help inform which features and what gems they’d like to see, as well as discuss with developers the potential for code contribution.

We propose a session to gather feedback from the community about which features would be of most interest and to also get a sense of a timeline that would be helpful to others. We will briefly describe some of the Avalon features and, as time allows, expand a bit on those that raise interest from those present. We intend to send out a survey with community suggestions on this after Hydra Connect.


Building a geospatial repository with GeoConcerns

Presentation - Eliot Jordan, Darren Hardy

Audience: All

GeoConcerns is a plugin to CurationConcerns for managing geospatial resources in a repository (http://geoconcerns.github.io). This presentation will give an overview and demonstration of GeoConcerns’ features and PCDM-based data model. In addition, we will discuss the code base and future development work such as integration with Sufia.

Audio: (MP3)

Case Study: Oregon Digital Migration from ContentDM to Hydra for digital collections

Presentation - Julia Simic, Linda Sato, Ryan Wick, Margaret Mellinger

Audience: All

This case study will address the initial decisions and reasons for switching to Hydra, prototyping for launch of Hydra head, metadata cleanup and asset migration, final quality review, and lessons learned.

Audio: (MP3)


DevOps group therapy

Breakout - Glen Horton and/or DevOps IG

Audience: All

Share with other participants how your institution is supporting and deploying your Hydra applications. What are your wins and fails? What's new or innovative in your Hydra infrastructure? What challenges are you facing delivering your application to your users? New Hydra adopters are invited to ask questions and learn from others.



Presentation - Yinlin Chen 

Audience: Developers, managers, sysops

In this presentation, we will present a software development and deployment process that we have developed at Virginia Tech. We have used this procedure to develop several Hydra projects for University Libraries services at Virginia Tech and have also used it to conduct interviews for Ruby and Rails developers. We implemented a central vagrant box to facilitate the development and deployment processes. Using this vagrant box gives us many benefits: 1. Code immediately. Our software engineers can focus on coding without worrying about software installation. 2. Unified development environment. All developers do their work in the same, consistent development environment. 3. Development of multiple projects simultaneously. Developers can work on multiple projects in parallel, switching between project environments in just a few minutes. 4. Minimal differences between development environments and production environment. After we finish implementation, we can quickly deploy our applications into a cloud environment (e.g. AWS and OpenStack) that is highly consistent with our development environments. We will present and demonstrate our hands-on experience on how we use a single vagrant box with different GitHub repositories to develop multiple Hydra applications in detail, including Sufia, GeoBlacklight, Fedora, Solr, Vagrant, AWS, and OpenStack.

Audio: (MP3)


Fedora 3 to Fedora 4 migration

Panel - Jim Coble, Adam Wead, Michael Friscia, Julie Hardesty

Audience: Developers and technically minded managers

Migrating a Fedora 3-based Hydra repository to Fedora 4 can be a major undertaking.  This panel will bring together representatives from multiple institutions that either are planning for, are in the midst of, or have completed such a migration.  Depending on where they are in the migration process, panelists will talk about their plans for migration, techniques used or expected to be used, obstacles anticipated or encountered, how problems were addressed, and migration outcomes.  It is hoped that the presentations and discussion will be useful to other Hydra sites planning such a migration.

Video Link: https://youtu.be/EpwRY62yFIs?t=6h19m32s


Fedora: Foundational Considerations

Presentation - Tom Cramer, Karen Estlund, Ben Armintor

Audience - All

Fedora provides an essential foundational layer to the Hydra Stack that may seem opaque to many in the Hydra Community. Recent community discussions have highlighted the  technical and community relationship between Hydra and Fedora, the value proposition of Fedora, and Fedora's role in an institution's broader preservation strategy. As the Hydra Community continues to thrive, the intersection of the larger repository community and role with these technologies is an important community rallying area.

This session is intended to: Generate awareness of the role Fedora plays in the Hydra stack; Generate awareness of the role client platforms play in shaping Fedora development; Advance a Hydra platform statement on the importance of supporting Fedora in general; Discuss methods for approaching development of features along the Hydra and Fedora stack continuum; Highlight the contributions many Hydra institutions are making towards Fedora as a way to advance their Hydra environments, as exemplars of how the two projects are symbiotic.

Audio: (MP3)


Geospatial metadata sharing with OpenGeoMetadata, and issues with modeling in RDF

Retitled!  See Sharing geospatial metadata using OpenGeoMetadata and Linked Data 


Google Scholar optimization 

Breakout - Sonya Betz, Weiwei Shi, Peter Binkley

Audience: All

Inclusion in Google Scholar is critical to recruitment of content for our institutional repositories - from our researchers' perspective, if Google Scholar can't find it, it doesn't exist! However, when the University of Alberta migrated its institutional repository to Sufia in 2015, we saw our Google Scholar indexing plummet from nearly 6000 titles to only 35. We've since made several changes to improve our results, and are finally seeing some success as Scholar slowly returns to crawl our content. We would love to have a discussion with other institutions about improving Google Scholar indexing for Hydra repositories. Come prepared to discuss any successes and challenges you've encountered, and we will share the strategies we've adopted to improve Scholar coverage of our Sufia repository.


Hydra: beyond the repository

Presentation - Julie Allinson, Jon Stroop

Audience: All

Using Hydra to manage and present cultural heritage resources raises a set of interesting challenges that are beyond the scope of the traditional institutional repository. These include more complex data models, elaborate and varied workflows, richer descriptive metadata, support for more and varied controlled vocabularies, the requirement to manage larger objects comprised of larger files and multiple derivatives, support for IIIF, and a desire for richer viewing environments in general. In this presentation we will discuss these challenges and highlight examples and implementations that have gone ‘beyond the repository’.

Audio: (MP3)



Presentation - Julie Allinson, Richard Green

Audience: All

In the UK, the Universities of York and Hull are looking at Archivematica's place in a research data pipeline. The two universities have slightly different use cases but share the desire to put research (and likely other) content through Archivematica on its way to the repository thus giving us a solid base for long-term preservation. We are both now in the third phase of a joint project to build proof-of-concepts to illustrate how Hydra and Archivematica can work together to manage and preserve research data. Since our project began, Jisc have launched an ambitious UK national research data shared service where a range of suppliers offer systems in different lots. Both Hydra / Fedora and Islandora / Fedora are part of the the ‘research repository’ lot of the service and the work of York and Hull has heavily informed the ‘preservation’ lot, with Archivematica one of the systems on offer. This presentation will describe the proof-of-concept work done by Hull and York, and will provide an overview of the new Jisc service.

Audio: (MP3)

Hydra-Stack / PCDM Live Walkthrough

Breakout - Mark Bussey

Audience: All, but there will be a technology focus

Using Sufia 7 as a reference model, what does the path for depositing a new item into a Hydra/PCDM based repository look like from an end-user, code, console, solr, and fedora perspective: show the upload of new digital content via the Sufia UI and show each step of the data flow in the Browser, rails console, and Fedora. Walk through derivative generation and show the various PCDM relationships being built at each stage. Then add a new user collection, and walk through the same flows as a work is added to a collection: what gems are in play, what does the data look like from a ruby developer perspective, what is persisted in Fedora and Solr - how does it all relate to PCDM?

Outline: google_doc

Video Link: https://youtu.be/EpwRY62yFIs?t=7h29m42s


Managing development projects

Breakout - Facilitator: Michael Friscia

Audience: Managers

A session where managers can meet to discuss approaches to adopting agile methodologies, balancing Hydra with non-Hydra project needs, inspiring staff, training, recruiting techniques, etc. We hope it will be an interesting place for managers to share their leadership styles/practices.

Managing Development Projects - Meeting Notes

Lightning Talks

The full list of Lightning Talks is here.


Migrating from MODS to Linked Data RDF

Breakout - Steven, Carl Anderson, Julie Hardesty

Audience: Not specified

Modeling complex application profiles in RDF without losing specificity is, well, hard. It is fairly uncharted territory regarding best practices for data concepts that don't exist in Dublin Core or DPLA MAP. Rather than just each institution making up their own unique / non-shareable way to handle these cases, the MODS to Linked Data RDF group was founded to get a consensus on how we would handle these less broad data concepts. This breakout session is to discuss current mapping choices the group has done, where institutions may choose to deviate from this "unofficial standard", work through mapping problems we may not have looked at thus far, and hear reports of any problems those trying to use this work have experienced in their Fedora 3 to Fedora 4 migrations. You don't have to have used MODS to attend - almost all of the data concepts are universal and we encourage participation by anyone looking to do more than Dublin Core or DPLA MAP!

Current mappings and group information can be found at: https://wiki.duraspace.org/display/hydra/MODS+and+RDF+Descriptive+Metadata+Subgroup



Breakout - Josh Gum, Linda Sato

Audience: Developers

A follow-up to our presentation at Hydra Virtual Connect to show the progress we've made on Opaquenamespace.org. We'll discuss how we are using Git and github as our master-copy for RDF graphs, and using Blazegraph and the triplestore-adapter gem for our operational datastore. 

Audio: (MP3)


Organizing regional Hydra meetings

Panel - Julie Allinson, Karen Cariani, Linda Newman, Joseph Rhoads, Chrissy Rissmeyer

Audience: All

Regional meetings are a great way to stay connected with your fellow Hydranauts throughout the year. Hear from past organizers about their experiences and tips, and learn about community resources to help get you started. It is also a chance to connect with other individuals in your region who are interested in organizing a future event.

Audio: (MP3 - partial, mixed quality)


Populating your repository with open access content from external systems

Panel - Linda Newman and others tba

Audience: All

A review of available APIs and services for identifying, and either manually or automatically loading, open source online content into a local IR.  Grey areas, policy questions, challenges and opportunities.

 Presentation Slides and Notes

Video: https://youtu.be/EpwRY62yFIs?t=5h8m42s


Profiling your stack - Everything's slow, what now?

Presentation - Trey Pendragon

Audience: Developers

The Hydra stack is large and complex, getting a handle on what's causing a specific slowdown can be difficult. This session would recommend some tools, strategies, and places to look for improving the performance of your application.

Audio: (MP3)


Report from the Sufia UI working group (PPT Slides)

Presentation - Michael Tribone

Audience: All

A recap of the group's recommendations for the upcoming now released Sufia 7 – thoughts on the diverse needs of the community and the result, What worked, what didn't work, how to approach it next time.

Audio: (MP3)


ScholarSphere Migration to PCDM - We Moved our data, you can too!

Presentation - Carolyn Cole

Audience: All

I will show the data model migration from Sufia 6 to Sufia PCDM we used for ScholarSphere. In addition I will the outline major design decisions we made along the way. Then we will look at the tools in Sufia for migration of data from Sufia 6 to Sufia PCDM. I will include examples of extending the functionality for people who have extended the basic Sufia 6 model.

Audio: (MP3)


Scoping and managing your Hydra project

Breakout - Shanita John, Mark Bussey

Audience: Managers

As a manager, one of the biggest challenges when starting any project (Hydra, or otherwise) is feeling comfortable that you’ve asked all the right questions before the first development hour is logged. But what are the right questions? And once you begin, how do you ensure that you, your developers and your stakeholders remain on the same page throughout the course of the engagement? This presentation will explore two processes that DCE employs to tackle these concerns; the Scoping Phase and the Inception Deck. We'll explore how these to practices work to consolidate project expectations and fend-off wild assumptions, helping you to feel confident that when everyone says ‘circle,’ no one is picturing a square.

Audio: (MP3 - partial)


Service management

Panel - Jon Cameron, Steve Van Tuyl, Debs Cane, Rick Johnson, Hannah Frost

Audience: All 

Service management is not just for managers! How managers, developers, librarians, and other team members can all work together to improve Hydra service management.  Sharing success stories and best practices; sharing challenges and potential solutions.

Audio: (MP3)


Sharing geospatial metadata using OpenGeoMetadata and Linked Data 

Panel - Darren Hardy, Chris Barnett, Andrew Battista, Kim Durante, James Griffin, and Jack Reed

Audience - anyone with a geodata interest

The session focuses on open approaches to sharing geospatial metadata. We will discuss issues around standard requirements, appropriate linked data predicates, using linked data for placenames and gazetteers, participating in organizations and networks for sharing like OpenGeoMetadata, and identifying available tools and resources.

Audio: (MP3)

Standard APIs and interfaces for interoperability

Breakout - Trey Pendragon

Audience: All

We've settled on standard solutions like LDP and PCDM within our community, but what does it get us? This session would focus on the current and potential benefits of the standards we've fallen down on, as well as identify opportunities for further interfaces and collaboration with institutions who may not be using Hydra.


Technical walkthrough of the Hydra-in-a-Box hosted platform  

Presentation - Erin Fahy, Chris Beer

Audience: Developers

Join us for a technical overview of the Hydra-in-a-Box hosted platform where we will detail how we're using Amazon cloud services and how we've evolved Hydra and Fedora to use them effectively. We will discuss how we've architected the traditional Hydra stack to be highly available, scalable, auto-deployed, and cloud friendly.

Audio: (MP3)


Using Questioning Authority to access Linked Open Data (LOD) authorities  

Unfortunately this presentation had to be cancelled.

Presentation - Lynette Rayle 

Audience: Developers

The Questioning Authority gem, which provides a set of uniform RESTful routes to query controlled vocabularies, was recently updated to support access to Linked Open Data (LOD) Authorities. This presentation will demonstrate configuring a LOD Authority, requesting a single term, searching for terms, and how to setup autocompletion for a field in a Sufia app. We may talk about caching strategies as well depending on the progress of this work which is under development.


UX Supercombo

Panel - Sonya Betz, Jenn Colt, David McCallum

Audience: Developers, managers

How can we include user experience [UX] best practices in Hydra's core development work? How can UX experts contribute effectively to the Hydra project? How can we ensure that decisions around elements that affect the user experience are based on good evidence? Through case studies and facilitated discussion, this panel will seek to demonstrate and explore how to better integrate end-user feedback into the Hydra development stream. Dave McCallum from the University of Oregon explores issues of accessibility and the user experience. Jenn Colt from Cornell University and Sonya Betz from the University of Alberta will facilitate an active discussion around building UX into Hydra development workflows, and will ask participants to suggest strategies for including real users and their feedback in design and development decisions.

Presentations: Accessibility(How) Can We UX-ify Hydra , Audio: (MP3)


What Does It Mean to Be a Hydra Partner?

Panel - Convener: Robin Ruggaber; Panel: Hector Correa, Julie Hardesty, Erin Fahy, Julie Allinson, Mike Friscia

Audience: All

What does it mean to be a Hydra Partner? What does the Hydra community expect of Partners?  What should Partners expect of the rest of the community?  Join us to discuss expectations, contributions and the advantages of Partnership.

Audio: (MP3 - mixed quality)