Searching the OPAC: The State of Play

1
Searching the OPAC The State of Play

• Peter Binkley
• peter.binkley_at_ualberta.ca
• Access 2007

2
Areas of Functionality

• Clumping
• Ordering
• Exploiting
• Contributing
• Deploying

3
Prehistory

• antarcti.ca (1999)

4
History 1

• Andrew Pace Making minor changes t library
  catalog systems is like putting lipstick on a
  pig. (LITA forum, Sept. 2005)?
• NCSU's Endeca OPAC (Jan. 2006)?
• WPOPAC (now called Scriblio Casey Bisson, Feb.
  2006)?
• Karen Schneider, How OPACs Suck, parts 1-3)
  (Mar.-May 2006)?
• LibraryThing XC (May 2006)?

5
History 2

• ILS vendors follow suit, led by Ex Libris with
  Primo (May 2006) Encore (III) EPS/Rooms
  (Sirsi/Dynix)?
• NGC4LIB listserv (June 2006)?
• Evergreen (Georgia PINES Sept. 2006)?

6
History 3

• Open source alternatives, notably Solr-based
• Casey Durfee, "Open Source Endeca in 250 Lines or
  Less" (Code4Lib, Mar. 2007)?
• Erik Hatcher / Bess Sadler, BlackLight (UVA, Apr.
  2007)?
• Andrew Nagy, VUFind (July. 2007)?

7
Clumping 1

• Bringing like together with like, for navigation,
  comparison and selection
• Notably by faceting on metadata fields
• But also by means of tags and other
  extra-metadata
• Examples
• NCSU (Endeca)?
• BlackLight
• Summa

8
Clumping 2

• Beyond faceting automated FRBRizing
• On the fly, based on online services
• xISBN (OCLC) (example)?
• thingISBN (LibraryThing)?
• Or systematically, based on algorithms applied to
  metadata
• OCLC Office of Research

9
Visualization

• Lists of facets can be processed into visual
  presentations as well as textual
• Literal geographical subject headings
• Example Peel Project
• Figurative abstract visualizations of various
  types, e.g. Aquabrowser

10
Ordering 1

• Traditional OPACs default to last-in first-out
• NextGen OPACs provide relevance ranked results
• But... does TF/IDF work for bib records?
• Examples
• U of Alberta
• Vanderbilt (Primo test)?

11
Ordering 2

• Problem of relevance ranking Libraries Australia
  solution (Dellitt and Boston 2007)?
• 1. Matches in the title, author and subject
  fields, and those fields which describe the
  format, nature or form of the item, are more
  important than general matches within the record.
• 2. Matches in multiples of the above fields
  are more important than matches in just one of
  those fields.
• 3. Where there is one or more query terms,
  an exact match of the term (where what was typed
  in to the query box is exactly the same as what
  is in a field in the record) is much more
  important than a phrase match (where what was
  typed in to the query box matches exactly a part
  of what is in a field in the record), which is
  more important than a word match (where all the
  terms in a query box appear in the field, but
  not necessarily next to each other).
• ...
• Example Libraries Australia

12
Exploiting

• Web 1.0 was the human-readable web Web 2.0 is
  the machine-readable web
• Make bibliographic metadata actionable
• Examples
• Export citations to e.g. RefWorks
• unAPI embedded link to raw metadata
• COinS embedded headless OpenURL
• Tools e.g. Firefox extensions
• LibX toolbar provides customized links
• Zotero citation manager

13
Contributing

• User-generated content
• Especially tags
• Example Tamworth, NH (Scriblio)?
• LibraryThing for Libraries import tags and
  recommendations
• Example Danbury, CT
• LibraryBiblioCommons approach (webcast)
• the fully social OPAC
• aggregated as broadly as possible

14
Identity

• An added dimension to tagging personal metadata
  is a potential source of facets, relevance
• Networks of trust
• OPAC is already an authenticated environment, but
  must interlock with other identity systems, e.g.
  courseware
• Shibboleth? OpenID?

15
Deploying

• New watchword Place our resources where our
  users live.
• I.E. courseware, Facebook, browser plugins (e.g.
  LibX), etc., VRE
• Interfaces SRU, OpenSearch allow mashups even
  OAI-PMH?
• Shall we pursue them as far as Second Life?

16
Issues and Obstacles

• Planning, determining specifications
• Metadata cleanup
• Back end processes

17
Planning and Specifying

• NCSU example eight months
• With one technical librarian, supported by a team
  and by the vendor (Anstelman, Lynema, Pace,
  2007)?
• Issues of scope what goes in? Is this only a
  catalogue, or something else?
• Example
• WorldCat Local

18
Metadata Cleanup

• Faceting exposes (bad) metadata
• Example BlackLight
• Do you know how clean your data is?

19
Back End Processes

• Extracting bibliographic and holdings records
• Enhancing them with external metadata
• Processing and Indexing
• Maintaining synchronization in real time (NCSU
  1.7 million records, updated nightly BlackLight
  3.8 million, similar)?
• Circulation status... (NCSU nightly)?

20
Services on the Grid

• Alternative approach scoped view of union
  catalogue
• e.g. WorldCat Local, or Talis Library Platform
• Or Google?
• Advantage broader aggregation of content
• Disadvantage complexity of including local
  holdings-level metadata, integration with local
  systems

21
Special Collections

• Benefits of providing access to collections with
  special needs
• e.g. Umlaut special handling of technical
  reports
• e.g. Blacklight faceting of musical instruments
• Escape from one-size-fits-all approach of OPAC

22
Imperative

• Users do not like our interfaces
• They do not use our advanced search features
• They do not get the full benefit from our
  collections
• We will only be left further and further behind

23
But...

• What if we solve the discovery problem for
  OPAC-contents, but it doesn't work for the rest
  of our users' citation-space?
• Integration with metasearch? OpenURL resolvers?
  (Umlaut)?

24
Continuity

• The technological opportunities change
• The service imperatives do not
• Traditional core values and methods of
  librarianship will see us through

25
Searching the OPAC The State of Play

• Peter Binkley
• peter.binkley_at_ualberta.ca
• Access 2007