SFX Implementation at Washington State University

1
One Stop Shopping (Almost)

• SFX Implementation at Washington State University

Sarah K. McCord WSU Health Sciences
Library Presented at the American Association of
Colleges of Pharmacy Annual Meeting July 14, 2002
2
SFX Implementation at WSU

• Definitions and Demonstration
• Technical Specifications and Support Issues
• Usability Testing
• Use Statistics
• Comparison with Other Products
• Future Directions

3
The SFX Team

• John Webb, Associate Director for Collections and
  Systems
• Janet Chisman, Systems Librarian
• James Czebotar, Computer Support Analyst
• Joel Cummings, Sciences Electronic Resources
  Librarian
• Ryan Johnson, Social Sciences and Humanities
  Electronic Resources Librarian
• Sarah McCord, Health Sciences Electronic
  Resources Librarian
• Scott Walter, Head, Education Library

4
Definitions and Demonstration
5
SFX What Is It?

• The Dream
• All Library Resources and Services Accessible
  From Within Any Library Database or Service!
• The Reality
• Many library resources and services are
  accessible from within some library databases.
• As more resources adopt the OpenURL Metadata
  standard, the Reality will approach the Dream.

6
A Word About Metadata

• Data About Data
• Not a new idea (remember card catalogs?)
• Why should we care?
• Knowing the description of a document allows the
  entire document to be manipulated
• Imagine linking your course syllabus to the
  library, a collection of digital images, relevant
  professional sites, and more--WITHOUT having to
  spend time creating web pages!

7
SFX Basic Vocabulary

• Common Terms
• Context Sensitive Linking
• Open URL Metadata Standard
• Source
• Services
• Target
• Thresholds

8
SFX Basic Vocabulary

• Context Sensitive Linking
• Uses metadata to generate a menu of options
  relevant to the starting point (Source).
• Open URL Metadata Standard
• Allows the SFX server to extract information and
  generate a menu of options.
• Source
• Starting point. Reference or citation the user
  would like more information about.

9
SFX Basic Vocabulary

• Services
• What the user can do, such as getting full-text
  or searching the catalog.
• Target
• Selected by SFX based on the Source metadata.
  (Also known as Where you end up.)
• Thresholds
• Controls established to limit potential Services
  to only those available to the user.

10
SFX Source to Target in ProQuest
11
SFX Source to Target in Web of Science
12
Technical Specifications and Support Issues
13
Server Specifications

• Operating system
• Red Hat Linux 6.1, 6.2 or 7
• Disk Space
• Minimum 3 Gb
• Recommended 10 Gb
• 2 Gb RAM, 1Ghz, Dual processor
• Cost
• 9,000 (today could get similar machine for
  6,000)

14
SFX Support

• Training two day, on-site
• Manuals
• User Guide
• Server Administrator
• Listserve
• One to One with SFX
• User group being discussed

15
Personnel

• Computer support analyst
• Setup and Upgrades
• Systems Librarian
• Input and update content
• Troubleshoot linking problems
• Electronic Resource Librarians
• Consult on services/display/usability/problems

16
Implementation

• Server setup early June 2001
• Two day training by SFX - mid-June 2001
• SFX team worked on setup, etc.
• Live to campus - mid-August 2001
• Publicity
• Classes and reference
• Librarian liaison contacts with departments
• More planned

17
Web-based Database Manager

• Activate sources, targets, and services
• Activate portfolios within targets
• Add local information to targets, services and
  portfolios
• Create local sources, targets, portfolios

18
SFX Server

• Configure the SFX services menu
• Dataloader for loading large files
• Export data from the SFX database
• Set thresholds
• what displays when

19
Data Logged

• IP address of user
• Data/time of SFX request
• SFX target requested
• Date/time at which SFX service is requested
• Type of SFX service requested
• Details of the Open URL used to deliver SFX
  services
• getFullTxt service available

20
Usability Testing
21
Usability Testing An Overview

• Designed to test software, not people
• Nonetheless, IRB approval was secured
• Random sample not necessary
• 3-5 participants adequate for test
• Rubin, Jeffrey. (1994). Handbook of usability
  testing how to plan, design, and conduct
  effective tests. Wiley New York.

22
The Testing

• Self-selected users of Web of Science or ProQuest
• Test subjects are given tasks and monitored by
  two observers
• Obsearvers record time taken and whether task
  completed (yes or no)
• Notes taken about test subjects behavior and
  spoken comments
• Exit questionnaire

23
Task 1 Using SFX, no instruction

• Task 1 Do you see anything on this screen that
  you could click on to find an actual copy of the
  article?
• Test Will participants use SFX button if they
  have never heard of SFX?

24
Task 2 Using SFX, with instruction

• Task 2 Participants are informed that we are
  testing software named SFX and told what it does.
• Test Will library users click on the SFX button
  after they have received minimal library
  instruction?

25
Task 3 SFX Services Menu

• Task 3 Starting from the SFX Services Menu,
  participants are asked to determine whether a
  journal is available at WSU.
• Test Will participants use the Search Griffin,
  WSU Libraries catalog option on the SFX
  Services Menu?

26
Task 4 Web of Science Only

• Task 4 Is there a way to find out how to get a
  copy of this article?
• Test Will participants use the Ask a Question
  service?

27
Task 5 View Full-Text vs. SFX

• Task 5 Locate this article from Cryogenics that
  you found in the Web of Science.
• Test Will test subjects use SFX, instead of the
  View Full Text button?

28
Exit Questionnaire

• If the SFX button had been labeled Washington
  State University Libraries, do you think you
  would have been more willing to click on it even
  if you had no idea where it might take you?
• Was the organization of the SFX menu
  understandable?
• How easy/difficult was it to use SFX?
• Was the wording on the SFX Services menu
  understandable?

29
(No Transcript)
30
Conclusions Improvement Needed

• Meaningful buttons in SFX source databases
• Full-text
• We chose Locate Document in FirstSearch
• Shorter labels for available SFX services
• Clearer citations on SFX menu
• Understandable phrase in the SFX menu banner

31
The Outcome
32
Use Statistics
33
Statistics Predefined queries

• Number of SFX requests per day
• Number of SFX requests per Source
• SFX requests sorted by Object type
• Journals accessed via SFX, ranked by use
• Books accessed via SFX, ranked by user
• SFX services requested, by Service type
• Full Text Targets, ranked by use
• Services preferred over full text

34
Average Number of Requests Per Day
35
Number of Requests Per Source
36
Top 12 Full Text Sources Requested
37
Top 5 Service Request Types
38
Top 10 Full-Text Targets Ranked By Use
39
The Competition
40
Other Cross-linking Options

• CrossRef
• Links from reference list of electronic journal
  articles to other full-text electronic journals
• PubMed LinkOut
• Links from within PubMed to full-text sources
• Library branding now available

41
Other Cross-Linking Options

• MAP - Innovative Interfaces Product
• Web Access Management
• Authentication System
• MetaFind - Universal Search Interface
• Allows searches in multiple databases
• Results in a single customizable interface
• WebBridge
• Links from the Catalog to other related materials
• Uses Open URL
• Context Sensitive Linking

42
Other Cross-Linking Options

• Endeavor ENCompass
• HTTP searching for web-enabled databases
• XML searching for structured requests and receipt
  of information
• Z39.50 gateway searching

43
Future Directions
44
MetaLib

• New Product from ExLibris, creators of SFX
• Promises Customizable Cross-Database Searching,
  in addition to Dynamic Linking
• Similar to Portals
• First course integrating MetaLib with course
  syllabus, content, and classroom activities will
  take place in Spring 2003 at Washington State
  University

45
MetaLib Sample Screen
46
Endless PossibilitiesOn Campus...

• Metadata Course Syllabus Dynamic Linking
• Once again
• Imagine linking your course syllabus to quality
  CONTENT from
• the Library
• digital image archives
• professional sites
• dynamically generated reading lists
• WITHOUT having to spend time creating web pages!

47
and Beyond!

• XML data harvesters now in beta, OpenURL becoming
  more widespread
• Additional possibilities for content integration
• push services for new articles, delivered to
  you based on your CVs metadata
• new ways to evaluate publications in online
  archives through easier/more productive online
  community interaction
• anti-plagiarism measures integrated into paper
  submission procedures
• and MORE

48
THANK YOU!

• QUESTIONS?

mccord_at_wsu.edu www.wsulibs.wsu.edu/hsl/sarah/aacp2
002.ppt