Next Generation Library Interfaces

ALCTS President’s Program: Breaking Down the Silos: Planning for Discovery in Library 2.0 – an ALCTS Midwinter Symposium

Marshall Breeding – title of his presentation: Next Generation Library Interfaces: Overview of concepts and a brief tour of commercial and open source products

My random notes from Marshall’s presentation:

Started with OCLC Perceptions stat – where do you start an info search? 89% search engines… library catalogs, 2%

usage of library websites is going down, everything else is going up … hmm…

aside – that makes a good case for sticking library content on blogs… users will find you that way

Crowded landscape of info providers on the web – google, amazon, us, etc…

Nobody has to go to a bibliographic instruction class to use Amazon… Nice.

Amazon is so easy to use – Marshall accidentally bought a book during a presentation, it’s so easy

Demand for compelling library interfaces:

urgent need for libraries to offer interfaces their users will like to use
move into the current millenium
search in line with how the current web works

inadequacy of ILS OPACs:

OPAC modules … failing to meet customer needs – it’s not really built for customers

Change is Underway! Lots of movement to break out of the current mold of library catalogs

Marshall hopes the back end will be redesigned, too, to be more modern

Next-Generation Interfaces:

redefinition of the library catalog – the word “catalog” is not a good one

more elegant presentation (think amazon)

more comprehensive info discovery environments
no longer enough to provide a catalog limited to print resources
digital resources cannot be an afterthought
systems designed for e-content only are also problematic
forcing users to use different interfaces depending on types of content becoming less tenable

federated search currently operates as a plug-in component of next-gen interfaces

web 2.0 flavorings:

strategic infrastructure + web 2.0
a more social and collaborative approach
web tools and tech that foster collaboration
integrated blogs, wiki, user reviews, etc
avoid 2.0 info silos – don’t have separate blogs, wikis, etc – make sure it’s integrated

2.0 supporting tech:

web services, xml apis, ajax, relevancy-based search engines, social networking tools and concepts

scope of the next gen library interface:
attempt to collapse silos or draw appropriately from each silo
unified user experience
single point of entry into everything
print + electronic
local + remote
locally created content

Functions and features:
Interface features/user experience:
simple point of entry – optional advanced search
relevancy ranked results
facets for narrowing and navigation
query enhancement – spell check, etc
suggested related results / recommendation service
enriched visual and textual content
single sign-on

Relevancy Ranking:
Endeca, Lucene do a good job
web users expect this! – the good stuff should be listed first
users tend not to delve deep into a result list
good relevancy requires a sophisticated approach

new paradigm for search and navigation:
users drill down through the result set and faceted browsing
faceted search – gives users clues about eh the number of hits, etc – it’s more like an online store’s faceted/guided navigation
more visual, has navigational bread crumbs

talking about boolean – walmart doesn’t teach their customers to do fancy boolean search to get to their products… we shouldn’t do this either!

Amazon doesn’t say “no results found.” Did you Mean and other features instead
validated spell check
have More Like This recommendation service
goal – make the query and the response to it better than the query provided

appropriate organizational structures:
LCSH vs FAST (faceted application of subject terminology)
full marc vs dublin core or MODS, or unstructured data
discipline-specific thesauri or ontologies
“tags”

enriched content – book jacket, summaries, etc

personalization/single sign on

deep search:
entering post-metadata search era
web searches full text. Google print, google publisher, open content alliance, etc
high quality metadata will improve search precision
commercial search providers already offer search inside the book
library search doesn’t do this!!!

Beyond discovery to fulfillment / delivery: this is the harder part – harder than discovery

Enterprise integration:
ability to deliver content and services through non-library apps
courseware, portals, social networking environments, etc

Great Benefit, Great Cost

We’re WAY TOO SLOW. Time on the web moves quickly! We need to catch up.

ideas to buy/use:

Endeca – one of the first
Widely used in the commercial world
high-dollar approach

aquabrowser:

LibraryThing for Libraries:
Wow – they are now distributed exclusively by RR Bowker

Primo: tailored for academic libraries

Encore from Innovative Interfaces (Nashville Public Library uses it)

Worldcat Local

TLCs LS2 (Shanandoah Public Library)
good visual design

SirsiDynix Enterprise
not aware of anyone actually using it yet
it’s a hosted product
does relevancy wel
uses chilifresh for book reviews
Marshall’s example is very ugly! Sirsi really needs a visual designer!

Scriblio:
Wordpress – looks great
Marshall’s not sure how it will scale
same stuff – faceted search, relevance, etc

VUFind:
production cat for the National Library of Australia – that’s pretty big.
open source, looks great

BiblioCommons
focuses on social networking
tag, review, comments, etc
oakville public library in ontario – in production.
Looks great!

Summon
serials solutions produst
eXtensibe Catalog

Polaris, Koha, Evergreen – doing well with providing next-gen features too

Q/A:

question/comment: we have a next-gen catalog, our faculty don’t get it – don’t understand faceted search, don’t know what a tag cloud is, etc – how do you get around that?

Answer: well, Amazon doesn’t seem to need to explain their faceted search, tag, etc stuff… ouch!

IL2008: Implementing a Next Gen OPAC

Speaker – Jeff Wisniewski

problem: old opacs weren’t designed for usability
– they were designed for the back-end of libraries
– designed to store data

Look at Jeff’s library’s OPAC – pittcat – much nicer looking than most other opacs (it’s still in beta)

they plan on keeping the more traditional-looking opac search for the forseeable future – eventually they’ll look at useage stats and go from there.

goals for new interface:
– does it have an intuitive interface?
– zero instruction needed, like amazon’s search interface
– no dumb error messages
– expose more of their collection (faceted search thing)
– integrate various silos (other databases they had, locally-created stuff, etc)
– get it up and running yesterday!

If you can, don’t do an RFP. It’s sorta backwards, hard to write, frequently states the obvious

instead, they made a features list
– must have
– highly desirable
– and a third category
– had to have the must have features

made a spreadsheet for the selection process with each vendor and their features listed.

Do this for your users – not your staff!!!
– boolean, advanced search, etc – your users DON’T CARE. So don’t offer it.
– resist the power search…

they renamed the old catalog the “classic” catalog – funny!

Do publicity on the new thing!

Usability
– call it what it is – map, not cartographic resource; music, not sound recording (that’s what the majority of them are)
– hyperlinks – you can put them practically everywhere

make sure it’s visual – use book jacket pics, etc

give your catalog legs:
– create facebook search widgets
– embed search widgets elsewhere, too

social stuff – do it even if you’re not quite ready yet

integration with other databases (federated search)
– some have them