Good morning Calgary. Day 2 of Access 2014.
My presentation was first up, and is fully written up in a separate blog post, We’re All Disabled! Part 2: Building Accessible (Web) Services with Universal Design.
When Campus IT Comes Knocking: A New Model for UBC Library IT in the 21 Century
- Paul Joseph Systems Librarian, University of British Columbia
Preambles: Library IT now integrated into central IT. IT doesn’t always understand profession and labels, called a business architect.
Why did it happen? Money. Streamlining and sharing of technologies and support. Three year budget reduction.
IT assessment involved 3 parts:
- business analysis broke stuff down into categories:
- develop, provide access, and assess & maintain collections
- engage: with teaching & learning, with research, with community
- manager: empower employees; internal knowledge & change; facilities, IT & finances
- Applications Inventory: 130+ applications, mapped to capabilities
- helps to orient priorities, what is unique to the library, mark ones from certain categories e.g. finances that would be better for enterprise level work
- very complex system
- infrastructure review: to do hardware inventory, application infrastructure, security issues
- establish governance
- sustainable staffing model
- leverage central services
- standardize where possible
- improve communication
- establish regular security activities
- resource / knowledge loss
- financial and investment planning
- data loss (data integrity)
- potential security concerns
- lack of scalability and flexibility
Cool latte art (not that we had any at the conference)
## Taking Control of Discovery: In-house Development to Improve Student Experience and Break Down Silos
* Sonya Betz Web and User Experience Librarian, Grant MacEwan University
* Sam Popowich, Discovery System Librarian, University of Alberta
Students are looking for books, articles, etc. How do they find them? Have all the usual tools, but comments include “search engine sucks”, “impossible to find”, “confusing”, etc.
Students feel really lost. Issues:
- too many confusing UIs
- responsive design
University of Alberta Discovery
Some extra issues:
- data silos e.g. knowledge bases
- customization, flexibility
- sustainable (capacity building, want to keep doing these things in-house)
- common infrastructure
- automation, interoperability
- who uses our website anyway? students are using google scholar, etc. want to actually know the answer
DAMS (Digital Asset Management Stack): OpenStack, Dataverse, Archivematic, Hydra (Fedora, Blacklight), Discovery: Blacklight (Solr, Rails). Single system. “Own your stack” -@adr
Using Ansible (deplay systems and services in automated fashion), Ruby, Vagrant (virtual box to spin up virtual machines), Scrum to write more tests, Rails.
MacEwan University Discovery
1st test case: iOS app to integrate vendor services in single interface, but only iOS, and different experience in mobile and desktop.
Developing a responsive integrated online environment (VSIP) where APIs go into a custom search API -> Custom Drupal search module = full control of the search UI
Similarly, custom drupal modules to integrate everything else e.g. room booking, eReserves.
Challenges and next steps:
- very complex & resource intensive
- can’t integrate everything, need to pick and choose
- project complications
- production launch hopefully in Jan
Beta launch in January. Usability testing to be done.
modules for placing holds, etc. close to single sign on as possible.
Development environment: Vagrantbox of base service image (written with kickstart), Ansible playbooks for components (e.g. MySQL) and applications (e.g. Discovery application)
Developer handbooks include:
- best practices, policies, procedures
- Vagrant (how to ddeply base image)
- Ansible (how to work with our playbook tree) similar to Chef/Puppet
Developers and sysadmins as a team, with not quite scrum, note quite devops, using birds of a feather sessions, variety of ways in communication (e.g. google chat, github issues) resulting in better communication, information sharing, etc.
- data: how do we break down silos?
- OAI-PMH (parse MARC to Blacklight
- Opinionated Metadata (OM) – allows specify vocabulary interested in, pass XML nodes, map XML records to metadata vocabulary you want
- Solrizer – how you want index, into Ruby, which is then indexed
- doesn’t solve all our problems
UI: bento box style search, content including hours, maps, subject librarians, libguides (maybe)
Industry shift towards vendors providing services instead of interfaces? e.g. Summon, Primo
Do our users still use the library website to search for, discovery, and access our materials? If not, what are they using instead and why?
How do we integrate library services with user’ workflow?
How do we provide data and services in easily reusable forms? Open APIs, Open Data. Let user build the interfaces and applications they need.
## Adding E-resources License Information to Library Systems: Three Libraries’ Approaches
* Jenny Jing, Information Systems Librarian
* Marc Lalonde, Web Coordinator, Librarian, University of Toronto
* Amaz Taufique, Assistant Director of Systems and Technical Operations, Scholars Portal
* Christina Zoricic, Metadata Management Librarian, Western University
OUR = Online Usage Rights
Events around Access Copyright had major impact to university library systems. Investigated available options, but available systems didn’t cut it. Ideal solution needed to be easy to implement, use, share; multilingual; cheap or free.
Consulted with layers to help legalese, librarians (bilingual, share it with users).
Started with UBC Library’s Mongo Licence Cruncher
Enhancements: different yes/no text, Ask text, OpenURL integration, security: SSL and lagout, attach PDF to license, APIs, activate/deactivate.
Embedded into SFX by mapping SFX and license database, but wanted to get away from having to do mapping, so moving to API.
At Western University
End of Access Copyright agreement in Dec 2013, and launch of Copyright @ Western to educate and inform for users to make informed decisions of using published works.
Licensing Project Plan: get all e-resources linked to license record, provide information about licensed use in various places
Major database cleanup, and e-resources not in ERM, so manual process. Have to maintain two knowledge bases (one for link resolver, SFX, and e-resources records).
At Queen’s University
Internal ERM DB, connects to OUR, then connects to link resolver (360 link).
Modified Voyager files by adding license link in holdings record.
Discovery system: used 360 Link API and assign license page.
University of Toronto
Local ERM for resources that don’t fit into Serials Solutions ERM. License module in Serials Solutions did not meet needs. Developed custom digital rights applications using Serials Solutions and EIR reports, allowed collection department to map them to licenses
Using APIs: Summons, Endeca, 360 Link (built own interface using API), OUR
UTL has db lookup table (because SFX uses different references) to work with OUR db.
### Improving Integration
Managing Mappings Dilemma, where potential solution is to use UofT created mapping tool, SP imported into OUR, using active targets (SFX), active DBs into OUR regularly with mappings or use API to query OUR for license information.
Have a jealous owl