Guest Post: Preparing for the Deep Digital Media Dive

Guest post authored by Frances Lloyd-Baynes, Content Database Specialist, Minneapolis Institute of Arts
Original post published on LinkedIn Pulse

When we started our journey to “improved digital media management” at Mia (Minneapolis Institute of Art) four years ago, we knew what we wanted (long term vision? tick). Our problem? Missing and inaccessible digital assets, with no way to get at and reuse them without making a whole lot of phone calls, wasting time and resources. (See Douglas Hegley, “The Mystery of the Missing Map and Other Digital Disappearances”.) We knew the assets were out there: they’d been created and maybe used once, but where the heck were they now? So our task was clear: get in a new DAM, fill it, serve it up, problem solved.  We made our plan, submitted our grant, and received funding to make our vision a reality. (Thank you, IMLS!)

Life is never that simple, though, is it? Filling the technology gap with a new DAM has been easy, but it is turning out to be the less impactful part of what we are delivering.

Yes, we needed a home for the assets being churned out by our designers, videographers, photographers, and educators. It had to be easy to use, straightforward, and integrate with our existing systems (an enterprise DAM, a works of art database, and website CMS). We chose an open-source DAM -  ResourceSpace (by Montala) - so we could access the underlying code and  tailor it to our needs. Our grant allowed us to bring in a developer to work on both ResourceSpace and the final piece of our project puzzle: a new search interface to unify all of our digital content systems.

ResourceSpace has (since June) simply and beautifully delivered a place to house and provide access to all of those wayward digital assets. Yet the new unified search interface is turning out to be the star of our project, with the potential to change significantly our digital media landscape. ResourceSpace makes finding assets easy; our other, enterprise DAM is less user-friendly. Our works of art database is both complex and has seriously restricted access. And none of the content systems on their own can provide a complete overview of the content we have related to our collections and our activities.  Enter MetaMia.

 

So very Meta

 

MetaMia is the new, browser-based interface we created to join all this content for one deep, inquiring dive. How did we do it? We extended our existing APIs (application program interfaces) to draw out the content from the source systems, then brought in ElasticSearch and Kibana to index and deliver that content up via browser interface.

In MetaMia, every word is searchable, all fields discoverable. Users can choose any or all of the four underlying systems to search, filter for asset types (video, audio, collection records, etc.) and for data mapped across systems, as well as use a field-by-field advanced search option. Assets are not only discovered, but can be directly downloaded from MetaMia, removing the need for our less-tech-friendly users to interface with our more complex systems. For a museum of (mostly) non-techies, it’s pretty heady stuff. (We received a spontaneous round of applause when demo-ing MetaMia at a recent all staff meeting.)

Early in the project, based on information we gathered from stakeholders, our team developed a set of guiding principles for managing the organization’s digital content. The core principles are

• Simplification (of processes; discoverability; use)
• Maximized resource (staff effort; asset use/re-use; system use/re-use; re: asset value)
• Flexibility (of formats; systems; use contexts; access/discovery; input/output)
• Clarity (clearly documented assets; standardized metadata; intuitive systems; museum processes & policies; user awareness)
• Independence (of formats; of users (i.e. they are capable); of systems)  
• Commitment (of the organization towards staff, support, resource)

With ResourceSpace and MetaMia, we are confident we have largely succeeded in fulfilling these principles. There’s still work to do, of course.

 

Making Meaning

 

Now that we have the search infrastructure, the pressure is on to offer up meaningful results. That means providing consistency (of content, structure, and language) across systems in ways we have not had to deliver before. As a collections data specialist, this is the stuff I get excited about: capturing that data, structuring it and controlling it to make meaning.

‘Making meaning’ requires getting the data into each system and associating it with the assets, into the right fields, and using consistent language. Here we have issues. We have significant gaps in the data we’re capturing - things we know people want to search by, but cannot because it’s not there now. Some of our systems have deeply complex metadata structures, others have virtually none. Some have built-in authority controls, others not. There is indeed work to be done to level the field.

The project is now drawing to a close and my final, personal goal, is to build a layer across all our systems for linking to common vocabulary controls. We are still working out how to deliver this, but looking at linked open data (LOD) to help. We need tools that sit outside individual systems, but which can be shared, e.g. the Getty’s vocabulary tools (Art & Architecture Thesaurus, Thesaurus of Geographic Names, and Union List of Artist Names) now being released in LOD format. And we need tools that can extend the meaning of our content (e.g. via translations into other languages) to a global audience.

Focusing on implementing a solution with the widest possible reach will help us move what is now an internal-only service to deliver deep content to a world of potential users. As it evolves, MetaMia will serve its assets directly back to our website, enabling the world to see the media-rich content we hold related to art, culture, history and our collections.
Mia is committed to the open source movement. Be assured that all our developments will be shared with the community via the Mia Github site.

If you would like to hear more about our project, my colleague Josh Lynn and I will be presenting at the upcoming Henry Stewart DAM Chicago 2015 event in September, where we’ll be sharing our work and its results. Note: As a reader of this post, you can register for that conference at a discount by using the code MINNEAPOLIS100. We hope to see you there.