Library Web Chic

Getting the Goods: Libraries and the Last Mile - Roy Tennant

The ‘Last Mile’ Problem

• Libraries have dealt better with description than delivery
• Our system are ill-suited for online delivery
• We have largely rested on our restrospective conversion and automation laurels
• We are being overtake by other players (ie. Google)
• We need the "two click solution" (Jeremy Frumkin)

• We have far too many clicks for our users

Books (Finding Fulltext)
Google - one click to search, one click to get

Melvyl Catalog - Search, get records but no link to book, another click to locations, and then scroll down, copyright information, then get to book

Google Book Search - Search, get results, but only shows it as limited access even though it is available for free

Mirlyn - Search, get to record, electronic locations listed, UMich version of book, Google Book Search it is only available via snippet view (not fulltext)
Google Books expects people to find books by searching the fulltext of the book and won’t show snippets unless you submit search terms.

Worldcat is adding records for all the book digitized in fulltext. They are also working on getting things for the right audience level.

Google connections to OpenWorldCat to tell people where a book is available nearby.
Libraries are going this way as well. "Find in nearby locations" Use OpenWorldCat to link to show what other libraires nearby might have a given book.

Catalog Availability Web Service - Tito Sierra, Emily Lynema

Will tell you whether or not an item is available or not. Or if a similar item (other edition, paperback) is available (uses xISBN)

Books: Key Compontents

• Google Books
• Open Worldcat
• FRBR concepts
• xISBN
• Audience Level
• Others?

Books: New Models of Access

• User goes to Google gets directed to libraries (sometimes); implication; we need to be ready.
• Libraries can provide better discovery of open content then Google is; implication: we need to collaborate
• Libraries need to fix out discovery systems; implication: clear your calendar for the next three years.

Journals

Hunting licenses (OpenURL resolvers)

Better Linking Through Chemistry

• Our goal: get the user to what the want as quickly and painlessly as possible
• OpenURL resolver are a beginning not the end
• Umalaut by Ross Singer

Umalaut

• OpenURL resolver is one piece of the linking puzzle
• Get better metadata to the start
• Looks at free agreggators of content to find fulltext
• Uses OCLC OpenURL Registry

GUF: Getting Users to Full-text

• Title links on results screen to lead to either:

• Fulltext

• Print holdings information with map

• Prefilled interlibrary loan request form

Issued Addressed:

• Get rid of the long line of clicks

Uses a number of different services

Journals Key Components

• OpenURL Resolver
• OpenURL Resolver Directory
• Other Resolvers (eg., DOI)
• Aggregators (eg arXive.org, oaister.org)
• Others?

It’s All About Intelligent Integration

• Many different services available to help get user to content
• Mostly we just need to combine them in useful ways
• Overcome challenges like vagaries of system we don’t control, bad, missing of misleading data

Library Chatbots in Electronic Reference - Anne Christensen, Hamburg University

What is a chatbot? - a computer program designed to simulate an intelligent conversation with human users. Many online companies such as IKEA use these.
Why a library chatbot?

• Students use libraries online rather than coming to library.
• Students use Google rather than quality resources
• Want students to be directed to resources which the library pays for

Creating Stella

• Developed a character distinct from the librarians. Not a librarian. Not really human.
• Software that governs how Stella interacts with people
• Perl with regular expressions

• phrases for different questions included in knowledge base

Integrate Stella into website using an Iframe

• movable, minmizable, can accompany users while they are browsing

Popular questions asked of Stella

• Finding books
• Borrowing
• Hours
• E-Resources

Stella confused as an interface for the catalog. She points out she is not the catalog and tries to help people figure out where to search.

Working on connecting Stella to the catalog

• User inputs information
• Search
• Stella gives back feedback on the number of the results, and user can see the results or user can get hits.

Working to continuously improve, especially initially. Have 15% are answered wrong. Don’t think that we are likely to get better results than that.

83% of dialogs on library-related topics

25% information literacy

Dialogs from Stella have helped to improve the library’s website. Learn about user behaviors, preferences, problem navigation areas

Other chatbots seem to be appearing in Germany.

Check it out (remember Stella is Germany-speaking) at http://www.sub.uni-hamburg.de/ 

LibX Firefox Extension - http://www.libx.org/

Students use search engines to find things because things like Google are easier to use.

Idea: a "library tool" for the Web

What does LibX do

• Direct access to the catalog and databases

• integrated toolbar
• smart context menu

direct access to open url resolver to find appropirate copy

Localization

• enriches web pages before they are displayed to the user
• inserts hyperlinked "cues" that point to a local library resources
• autolinking for known identifiers

30 live editions (Libraries using)

Downloaded and installed over 4,000 times

LibX you can download it, configure, compile, and install it or get one from the LibX stuff.

Working on creating an edition builder web interface

Want feedback