Scholars Portal's eBook platform
Presented at TechKnowFile 2011
May 27, 2011
What is Scholars Portal (SP)?
Scholars Portal provides a shared technology infrastructure and a collection of collaborative services for Ontario’s 21 university libraries.
The Scholars Portal team consists of 10 technical staff, 10 librarians, two students, and a team lead.
Some Scholars Portal Services
- interlibrary loans service based on the Z39.50 protocol
- repository of over 20 million articles
- web and mobile bilingual interface
- repository of over 360k books in electronic format
- Allows users to search, tabulate, and analyse statistical data sets on the web
- searching data from dozens of different statistical agencies and polling companies
Works in Progress
- captures, stores, analyzes, manages, and presents data with reference to geographic location data.
- includes an integrated mapping tool
- intended as both a research and a literacy tool
Homogeneous Search API
Mobile versions of services
- Books is in progress, mobile UI currently available for Journals
Personalized services / user accounts
- bookmarks, notes, highlights, personal library, RefWorks integration
- Use existing accounts
- School’s authentication infrastructure (Shibboleth)
- OpenId (Janrain Engage aka RPX)
- Easy integration with various SP applications
- avoid multiple shibboleth service providers
- work with different platforms
A project by Internet2:
“provides cross-domain single sign-on and attribute-based authorization”
- Used by many academic institutes. Canadian Access Federation in Canada
- Three components:
- Browser (user)
- Service Provider (us)
- Identity Provider (schools)
- Attributes we requested:
- persistent-id : used as persistent identifier to link returning users to their account
- email (encouraged) : used to provide alert service to users
- affiliation (optional) : used to provide differentiated service for various kinds of users (faculty, staff, etc…)
- common name (optional) : used to fill user profiles
Janrain Engage (RPX)
- Abstracts authentication for Google/Yahoo!/MSN/Facebook etc.
- Imports profile info and provides unified API to access them
- Include the widget in your login page.
- User will be redirected to return URL after authentication.
- Extract the token parameter from the POST data.
- Use the token to make the auth_info API call:
- apiKey: c04ad3eddbf6dbc918e9e0f5b36b5320d0e63
- Token: The token value you extracted above
Parse the response and extract the identifier. A sample JSON response:
'email': '[email protected]',
'preferredUsername': 'Chunky Bacon' } }
- “Exposes the majority of RefWorks functionality in a REST format”
- Divided into classes that handle specific functionalities:
- Attachments, Authentication, Authors, Descriptors, Periodicals, Folders, Batch, Deleted, ImportFilter, Manuscript,
MyList, OutputStyle, Properties,PubMed, Reference, Retrieve, RSS, SavedSearch, ShareProperties, Subscriber, SubscriberPrefs, User, Utility, Z39
- We use Spring’s RestTemplate to access RefWorks API on the backend
Controller extract RWSessionId from session and uses restTemplate to access:
- XML response or error codes are translated to a json response.
- Json response is returned to client.
- Single sign-out is not really working
- Proxies considered harmful!
- URL rewriting
- Domain change
- Test/Debug is not easy
- Provide a test page for sys-admins
- Privacy = no email address for you!