Panel of 3 on linked open data.
Catelynne Sahadath – A Change Would Do You Good
Back of house need to work together with front of house. The link is very important in order to properly serve users. Communication is key.
We stop ourselves about communicating effectively. We need an open line of communication.
Current change management is focused on communication, where empathy is a big part of it for everyone we interact with: other staff, admin, users.
Little literature in libraries, but vast amount in medical field.
Fake it until you make it:
- get away from your unfeeling inbox
- listen. If you’re talking, you’re doing it wrong.
- Ask a question.
In Real Life
- participatory management
- invite people with ticketing system, training Thursdays, informal interaction
- see others in their natural habitat (conferences, training, shadowing)
MJ Suhonos – The Way Forward is Open
Openness movements: open access, open source. Doing a good job of talking about them.
The two new forms of openness.
Libraries have not done a good job of getting involved in the internet. We frequently have little to do with the way thing work in the internet, which is a problem because it’s a big place.
We can reassert ourselves.
Open Data – Openness Outward
Data that people can take and do something with it, which is explicitly stated. This is asserting control because stating what others can do.
Open data is a policy framework. Clarifies what people can do with it. Allows a spectrum of licensing options.
In 2010 while working at TPL, 2.5 million MARC records submitted to the Internet Archive. This was a 9 month effort.
The coolest thing to do with your data will be thought of by someone else.
Built visualizations based on the data (and SFPL), which can actually be used by libraries for collection management.
Openness Inward – Linked Data
Uses the same technology as the WWW (URIs for names, HTTP for tretival), decentralized, open standards. Flexible enough to be used for libraries and the rest of the world at the same time.
Linked data is an accessibility initiative for computers.
Can begin to share and playing in that much larger sandbox.
Technical framework for data interoperability. A common language for sharing data online. Unrealized potential due to incompatibility.
The problem we had was that we didn’t talk to anyone outside of libraries, but now with RDF and linked data, we can do that.
LODLAM (Linked Open Data in Libraries Archives and Museums) is especially showing a big effort in Europe and Australia.
Bibframe’s goal was to replace MARC by Library of Congress. It’s an experiment where we can be part of the conversation.
Linked Open Data (semantic web is the intersection of linked data and open data).
- foster discussion around rights management issues and open data
- consider the impact of license restrictions when negotiating contracts
- establish institutional policies for data sharing and licensing
The status quo is broken.
- stop using proprietary vendor technology
- start using web based standards
- choose options that embrace openness
- learn about what others are doing and participate in community initiatives
LOD is empowering our libraries.
We can let other people solve some of the problems, and simply share in the solutions.
Mark Jordan – Let’s Do This Thing
We have the tools in our grasp now. Just need to think more imaginatively in how to use them.
We need to stop talking and start doing (more) (and continue talking).
Data is out there for linking
- LC linked data service – id.loc.gov
- Canadian subject headings
- Getty vocabularies
- DBpedia – out there and huge
DBpedia = queriable version of wikipedia
Through DBpedia, we can pull VIAF info into our catalogue.
Spotlight can great links back to DBpedia simply based on the text available.
All automatable, and all possible in real time.
** linked data to define relationships of items with internal / external items** using linked data authorities** embedding linkable data in HTML using schema.org and RDFa
- General Library Community
** paying vendors to clean up authority headings in our records
** using ILS that don’t effectively support genuine, uniformly maintained authorities
** using data structure (MARC) that are capable of supporting value URIs (see Karen Coyle’s blog)
** caught in the vicious circle of waiting for vendors to support better data structures (triplestores)
Things We Can All Do
- look for opportunities in our own communities and between communities
** understand and advocate on intellectual property issues surrounding linked data
- push developers, vendors, administrators (in the right directions)
- think linking
We can do this thing.
2 thoughts on “BCLA 2014: Linked Open Data and Libraries. When? Or NOW!”