Ronan Kennedy; NUI, Galway LIR Annual Seminar 2006 Needles in Haystacks : Metasearching and Management of Miscellaneous Material.

Ronan Kennedy; NUI, Galway

LIR Annual Seminar 2006

Needles in Haystacks : Metasearching and Management of Miscellaneous Material

Quick introduction

NUI, Galway experience

MetaLib in action

Metasearch workflow

Google Scholar

Information Literacy

Points to consider

Introduction Also known as:

Multisearching, cross-searching, broadcast searching, integrated searching

Metasearching is a process in which a user submits a query to numerous information sources simultaneously. Each of these resources has its own search engine. The metasearch system transmits a query to the search engine and directs it to perform the actual search. [Tamer Sadeh, Ex Libris]

A metasearch system ‘piggy-backs’ several other database search engines to search them at the same time. The results are grouped together and displayed in one group

Introduction Z39.50 protocol

Approved by NISO in 1988 “Z39.50 makes it possible for a user in one system to

search and retrieve information from other computer systems (that have also implemented Z39.50) without knowing the search syntax that is used by those other systems” http://www.loc.gov/z3950

The NUI, Galway Experience MetaLib from Ex Libris

Used in conjunction with SFX Link Resolver

Went ‘quietly’ live February 2005

Subject groupings to help usability

45 out of a possible 137 resources are cross-searchable

NUI, Galway’s MetaLib

A Quick Metasearch with MetaLib

A Quick Metasearch with MetaLib

A Quick Metasearch with MetaLib

A Quick Metasearch with MetaLib

A Quick Metasearch with MetaLib

Behind the scenes - metasearch

Interface A

Interface B

Interface C

Interface D

Interface E

Resource A

Resource B

Resource C

Resource D

Resource E

Z39.50 protocol applied

Search syntax translated

Results gathered by metasearch

engine

Grouped results displayed

With a metasearch engine & a link resolver…

Users don’t have to know where to go to begin a search Users don’t have to know what resources to perform a

search in – subject categories Users can save a massive amount of time searching Users can compare results of several resources

simultaneously

Users don’t have to know where to go to find the fulltext of the article

The link resolver saves time by ‘building a bridge’ to the articles

---- knowledge of Library resources no longer necessary

Stumbling Blocks…

Google Scholar Information Literacy

Google scholar

Google Scholar With a unique ranking process, unparalleled hardware

resources, sophisticated crawling techniques, and access to published materials, Google Scholar is being positioned as an essential resource [Tamer Sadeh, Ex Libris]

Google Scholar is already proving that a harvested, centralized search approach is more useful [the digital librarian blog]

If Google Scholar becomes a better provider of scholarly articles and information than a typical university library, then we’re going to struggle to justify not only our budgets but our role in the academic process [the digital librarian blog]

Google Scholar is good because…

easy to use the branded look and feel excellent search algorithm extremely fast no authentication issues covers a large broad range of resources

… but keep in mind … At what level does it make most sense for resources to be

aggregated for more effective use. Think of two poles – the fractured response available to a library user, and Google. [lorcan dempsey’s weblog]

Roy Tennant: it’s a good principle to centralise when possible

like Google Scholar – but that ignores the need to segregate when possible too. To use WorldCat as an e.g., a user may only wish to see items they can get their hands on. Google has no plans to offer any criteria to segregate. A central argument against Google Scholar – finding good stuff is as much what you don’t search as what you do

… and it’s bad because does not disclose information about its content, or

define scholarly some big publishers e.g. Elsevier, ACS and Emerald

are not yet included updates not frequent enough material that Google Scholar incorporates from a

publisher does not always provide complete coverage Google Scholar ‘plays fast and loose’ with hits (Peter

Jasco) (Google Scholar arranges results by relevance, taking into account the number of times that the items has been cited in scholarly literature)

Relevance – What is defined as scholarly?

Mercury – Freddie Mercury; Venus – Venus Williams

Google Scholar Vs Dedicated Metasearch Engines?

---- no contest!

Roy Tennant: “Only librarians like to search; everyone else likes to find”. Google Scholar has taught us, quite powerfully, that the user just wants a search box. Arguments as to whether or not this is ‘best’ for the user are moot – it doesn’t matter if it’s the best if nobody uses it

[‘Federated Searching: Put In Its Place’; Miller, Todd; LibraryJournal.com; 15/04/2004]

What students want: to save time, fewer choices, easy to use

Google-like interfaces, good enough results, fulltext [‘Brick and Click Libraries Symposium’; Cox, Christopher; Library Hi-Tech News; Vol 22 Issue 10]

But who really cares?

Metasearching & Information Literacy To be able to search so many different resources, a

metasearch engine has to sacrifice something ….…. Advanced searching e.g proximity operators, value-added features e.g saved searches

Some claim that libraries should take advantage of metasearching to serve patrons content with a ‘good enough’ answer. Good enough pretty much sums up metasearching quality, but should our schools and colleges be satisfied with mediocrity? Do we pride ourselves with producing good enough library researchers and good enough graduates? [‘Do Want Or Need Metasearching?’, Frost, William, Libray Journal 1/04/2004]

Negative impacts of metasearching on Information Literacy

The value-added features of a database are now lost – alerts, saved searches, proximity operators etc

No more critical evaluation of databases – just click them all and hope for the best

Users become less aware of our resources:“students do not know what resources we have, have little interest in learning about alternative titles … feel overwhelmed … lack the time and inclination to learn more” [‘Metasearching in Boston College Libraries’; Tallent, Ed; New Library World, Vol 105, No 1196]

What we have to work against :-Are we sacrificing learning skills for a convenient search interface? Does metasearching adovcate google-type thinking? Does it discourage advanced information literacy skills?

Issues for training Should libraries focus on training users to use native

interfaces more effectively, or use portals to increase overall usage?

More effective searching – less results but of a higher quality – less risk of information saturation

Can we separate users’ needs – undergrads, postgrads, researchers – metasearching will have a separate impact on each one – training implications

Separate strategy for non-compliant resources?

Information Literacy Standards ACRL Information Literacy Competency

Standards for Higher Education (http://www.ala.org/ala/acrl/acrlstandards/informationliteracycompetency.htm)

determine the extent of information needed access that information effectively and efficiently evaluate information and its sources critically

‘Information Literacy For The Real World’; Terrell John; Lifelong Learning Conference, CQU; 2004 [http://lifelonglearning.cqu.edu.au]

Information Literacy Standards ACRL Standard 1 – determine the extent of

information needed(amount and type)

Grouping several database types together makes it difficult to distinguish between those types & increases threat of information overload.

results may highlight more types than were originally anticipated – journal articles, standards, proceedings etc.

Information Literacy Standards ACRL Standard 2 - access that information

effectively and efficiently

Metasearch engines sacrifice ability to do complex searching. Users cannot do advanced searches to the same extent as on native interfaces.

Efficiently accessing the information is taken care of, and is aided by link resolvers

Information Literacy Standards ACRL Standard 3 – evaluate information and its

sources critically

Some native interfaces allow clustering of results

Having multiple resource results listed together does allow for critical evaluation of sources

Advantages of metasearch systems

Can search multiple databases concurrently [time and ease of use]

Related databases can be grouped per subject Only one interface – less learning required Automatic removal of duplicates (allegedly)

A locally controlled and branded system that enables librarians to control levels of access

For Librarians: increased usage of expensive/obscure databases

Disadvantages of metasearch systems

Compliancy! 45 out of 137 in NUI,G Metasearching is costly – hardware,

maintenance, training Information Literacy implications Problems with relevance ranking De-duplication technology not perfect Slow Risk of too many results – evaluating results more

difficult No database specific search features – applying

limits etc

What does the future hold? Technological developments: proximity operators,

thesauri, alerts etc

More compliant vendors

Improved speed

In Conclusion - for the moment… Metasearching will not help experts find information

with greater precision. It’s great for the average user – speed and simplicity For information experts it’s an excellent starting point

for evaluation & source identification

The Librarian’s Role? Accommodate everyone’s needs!

…… who knows what future metasearching developments will hold??