Professional Documents
Culture Documents
Jason Vaughan Director, Library Technologies UNLV Libraries Tamera Hanken Director, Technical Services UNLV Libraries
Services
y Part 4: Quick Tour of the Current Marketplace y The Big 5 y Similarities and differences
considerations
y Record loading and mapping (catalog content) y Harvesting and mapping digital/local content y Working with central index data (internal & external content) y Web integration and customization y Assessment and continuous improvement
Preface
And Today . . .
Web Scale Discovery Services
10
Federated search Metasearch Next generation catalog Discovery layer Web scale discovery
11
Part 1
12
agreements) & technology that greatly facilitates the discovery and delivery of a tremendous amount of purchased, licensed, and free information.
y [A service] that is delivered on demand to library
users via the browser, with infrastructure, processing and indexing provided and maintained remotely by the vendor.
13
14
their Public Domain eBook Collection; digitized public domain journal issues, etc. DOAJ - Directory of Open Access Journals (over 3000 journals searchable at article level) Hindawi Publishing Corp. (200+ open access academic journals) arXiv e-prints (Cornell Univ hosted, over 685,000 items mostly in the sciences) Some can incorporate and expose the bibliographic and digital collections of other libraries the collections that you yourselves have and have given permission to be harvested and discoverable.
15
text content for indexing purposes y Agreements may be brokered whereby the publishers provide fielded metadata (title, author, publication info, etc) to the discovery service vendors y Vendors can develop multiple content streams for the same, finite content. For any given article, there are lots of potential sources for that exact same article, not just the original primary publisher . . .
16
the licensed, full text content -- is still dependent on the publisher / aggregator content licenses the local library purchases / maintains. In some cases, the final result may be a citation/abstract information, such as is found in A&I indexes
y Still, you may have access to some citation level content which you
otherwise wouldn t have access to (and haven t licensed) even this is helpful for discovery.
17
tools to broker the access to the full text content y Link resolvers y Proxy servers y Other rights management knowledge databases associated with the discovery vendor
18
collections
y Content from other hosted repositories, such as
19
A single search box Faceted searching Evaluative content (book covers, reviews, etc.) Social networking tools, etc.
21
systems allow for local hosting of the interface (the content index is always remotely hosted in the cloud)
y Discovery Services are quite open compared to old-
school ILS platforms with flexible APIs and customization capabilities allowing you to hack, repurpose, or customize the interface.
22
Part 2
23
24
-- Marshall Breeding, Building Comprehensive Resource Discovery Platforms. Smart Libraries Newsletter, March 2011.
33
-- Marshall Breeding, Building Comprehensive Resource Discovery Platforms. Smart Libraries Newsletter, March 2011.
34
But to emphasize, of all the perspectives, the user perspective should not be underestimated . . .
36
In Short, Before . . .
37
Before . . .
38
Before . . .
39
40
41
42
Initial Questions?
43
Part 3
44
Evaluation
As history has shown, multiple solutions arise to address real needs, and each solution has its own characteristics. In terms of discovery solutions, I'm confident that each library, after conducting a thorough evaluation of facts and features, will be able to determine which of the available products best fits the library's mission, needs, policies, and environment. Nancy Duskin. Ex Libris Responds to Interview by Jane Burke, The Charleston Advisor. July 2010 12:1.
45
Evaluation
Acknowledge that y A new discovery service could be the primary entre for a majority of your users certainly your undergraduates to both local library materials and your huge portfolio of licensed e-content.
y Whichever service you choose, while, not permanent,
Evaluation
Unless you live under a repressive and controlling dean/director, or have a very rapid timeline in which you need to spend a lot of money, you may want to:
y Research, in detail, the (changing) marketplace y Be inclusive, and communicative, with your fellow library staff (and perhaps beyond) y While your institution may be unique, it may not be as unique as you think, so don t recreate the wheel y Don t rush to a selection, yet don t get caught in indecision, which, of course, is a decision
47
Evaluation Models
Published research can help. Examples include:
y Oregon State University Discovery Services Task Force Recommendation to University Librarian y University of Arizona Implementing Web-Scale Discovery in an Academic Library y University of Michigan Article Discovery Investigation y University of Minnesota Discovery Phase 1 and 2 Reports y University of Nevada, Las Vegas Investigations Into Library Web Scale Discovery Services
48
Evaluation Models
y CONTENT (scope and depth, richness, update frequency, ease of incorporating local content) y SEARCH (interface simplicity, quality of results, ability to customize relevancy, etc.) y FIT (ease of implementation, compatibility with existing
software/content environment, overall customer support, etc.) as justified vis a vis the libraries goals/objectives)
y COST (as a new service to existing tools, instead of other finding tools,
-- Luther, Judy & Maureen Kelly. The Next Generation of Discovery. Library Journal, March 15 2011.
49
references, etc.)
50
Implementations (late 2009 mid 2010) Question list to Vendors (Fall 2009) Task Force Presentations to Library Staff (Fall 2009, Spring 2010, Fall 2010) Library Staff Surveys (April 2010; June 2010) Content Overlap Analysis (May 2010+)
Summer 2010) Follow-up question lists, conference calls to vendors (Summer 2010) Detailed Q&A and conference calls with early adopters of these services (Fall 2010) Final Administrative Discussion & Recommendation (Winter 2010) Purchase (end of 2010)
52
environment by describing our various locally hosted and remote licensed content
y We developed and organized a list of 70+ questions
which we sent to all five vendors (whether they had a released product, or just an announced product)
53
Background Research
y Organized questions into nine broad areas, some
all did
54
Metadata y Section 3 Publisher/Aggregator Coverage (Full Text and Citation Content) y Section 4 Records Maintenance and Rights Management
55
Background Research
y Section 5 Seamlessness & Interoperability with y y y
Existing Content Repositories Section 6 Usability Philosophy Section 7 Local Look and Feel Customization Options Controllable by the Library Section 8 User Experience (Presentation, Search Functionality and What the User Can Do With Results) Section 9 Administration Module & Statistics
56
Staff Education
y Created staff website; library wide internal presentations y First Presentation
y
Education
y Second Presentation:
y Third Presentation:
y
multiple choice, and free text response questions y Respondents could skip any question they wished y Higher response rate from first survey; lower response rate for second survey
58
Staff Survey 1
y First Survey: Conducted BEFORE Vendor Visits. We asked questions in three functional areas. y Local Library Customization Capabilities.
y
Is it important for the library to be able to control/tweak/influence the following design elements . . . (Strongly Agree < - > Strongly Disagree)
y End-User Aspect: Features & Functionality y The following functionality is important to have in the discovery service . . . (Strongly Agree < - > Strongly Disagree) y Content
y
Please rank on a 1-10 scale how vital it is that a discovery service accommodate records from these information repositories
59
Staff Survey 2
y Conducted AFTER All 5 Vendor Visits/Demos y Had questions along the same functional areas as first
survey (local library customization features; end user features/functionality; content) y For each question, respondents were asked to respond to the question for each of the five products.
y e.g. The Discovery Platform appears to ADEQUATELY
cover a MAJORITY of the CRITICAL publisher titles (strongly agree, agree, disagree, etc.)
60
More Work . . .
y Collection Overlap Analysis y Consolidating Vendor Responses y Vendor Onsite Visits y More Questions for Vendors y Reference Checks with Early Adopters y More Research: Keeping Ahead of the Curve y Vendor Quotes y Final Recommendation to Library Administration
61
How long was the implementation period? Is it now your default search tool? Have you observed any particular strengths in terms of subject content in any of the three major overarching areas -- humanities, social sciences, sciences? Have you observed that the discovery service leans toward one or a few particular content types?
y Content Questions
y
with the discovery service s interface? Is there any particular feature or function that is missing or non-configurable within the discovery service that you wish were available?
62
y y
Interface design (real time status calls for ILS items; faceted navigation; advanced search; etc.) Content inclusion, local and subscription Customization capabilities, APIs, etc.
64
Part 4
A Quick Tour of the Marketplace (Similarities and differences between some of the services)
65
66
67
content (hundreds of millions of indexed items . . . at least two vendors indicate they have already surpassed a half billion indexed items).
68
their metadata and their competitor s metadata. You ll have to talk to them.
69
modes) y Faceted navigation (subject, content type, publication date range, etc.) to help users drill down a large set of results y Inclusion of enriched content such as book cover images y Shopping carts to easily mark items and later export the materials (email, print, save)
70
call number, location, and status information for library hardcopy materials
y Did you mean? spell checkers y User configurable RSS feeds to easily re-run searches
later
71
abilities
y Some systems offer tighter integration to the full text. y Some systems may offer more full text content as
native PDFs, which are of higher quality and searchable (as opposed to scanned image pdfs).
72
have an ILS from the same vendor; or if the journal content you re looking at is sourced from the discovery platform vendor
73
community features
74
y Algorithm Tweaking
75
76
77
harvested
y Whether you host the application or they host it y University FTE count and/or degree granting status. y Size of your user community
78
Vendor brokered content enrichment services Article recommender services Optional federated search components offered by the vendor Consulting / development of custom ingestors to harvest unique, non mainstream local library databases
79
Part 5
80
81
Things to Be Aware Of
y Does not cover 100% of your resources y Potential role of federated search y Can lose the unique interface / functionality of
82
83
parent company whose business is content . . . Is that parent s company content promoted or weighted more heavily in search results?
y In the sense that some vendors may be inking exclusive
agreements with publishers whereby only that vendor s discovery tool can index that publisher s content
86
88
Session Wrap Up
Part 2 Evaluating and Implementing Web Scale Discovery Services in Your Library July 20, 2011, 2:30 PM Eastern / 11:30 AM Pacific
Questions . . .
89
jason.vaughan@unlv.edu
tamera.hanken@unlv.edu
90