Classification of Resources IV: Information Architecture, Digital Libraries, and Social Classification

Subject Access Problem, Revisited

• How does this apply to Web searching?
• Automatic indexing (earlier in semester), such as what Google docs: Does a centralized index address the soda versus pop controversy?
• Today, we explore possibilities that center on the human organizing of Web resources.

Challenge of Providing Access to Web Resource Collections

• Can we build true digital *libraries*?
• Can we systematically arrange Web resource collections by subject in a way that is useful to those seeking information? 
• Can we replicate the "great reduction" phenomenon?
• Can we insert these digital libraries into the lives of our users where the time available for finding information is like to be a major constraint?
• While at home
• While at work

Search vs. Browse of Info Resources

• Search
• Of surrogates
• Bibliographic
• Cataloging
• Verbal subject analysis
• Results list
• A website's search engine
• Browse
• Of actual info resources
• Bibliothecal
• Shelf arrangement
• Classification
• Navigation
• A website's architecture

Organizing Hyperlinks to Web Content

• Hyperlinks to what types of files?
• Text
• Audio
• Video
• Powerpoint
• etc.
• Hyperlinks to whose content and to what kind of structures?
• Content: Our own content ("stickiness") or someone else's content (traditional library approach)
• Structure: Websites, sub-sites or webpages

Classification & Hyperlink Organizing

• Main goal is to transcend the limitations of shelf arrangement in physical libraries:
• E.g. multiple class numbers in classified catalog
• Arranging links in multiple hierarchical locations (poly-hierarchy)
• Another important goal is to take end-user perspective into account:
• "Views" of a website in the case customized for "Future Students", "Current Students", etc. 
• Social bookmarking (let "pop" be pop)

What is the Documentary Unit?

• For collections of hyperlinks to internal Web resources (i.e., self contained websites):
• Keeping users on the webpages of your website
• The emerging field of Information Architecture
• Also referred to as navigation design for self-contained websites with few links to external content.
• About creating navigational and organizational structures that put users in touch with the information they need in a website as efficiently as possible (similar to physical library signage, but also vocabulary control for labels).
• Job titles include:
• Interaction/Interface Designer
• Usability Engineer
• User Experience Designer (UX)
• For collections of hyperlinks to external Web resources:
• Linking to webpages: Google and other search engines
• Linking to web sub-sites: Human constructed link indexes such as those on library websites
• Linking on entire websites: Human constructed link indexes such as those on library websites.

Structural Standards for Web Resources

• Are websites structured link books? No. Not always. That's why we have to worry about the differing structuring of information. 
• There is a lack of standards for structuring websites across publishers:
• Many "vanity" web publishers
• "Commercial publishers are making inroads

How Libraries Organize Web Resources

• Standards approach: Cataloging - use of 856 tag:
• MaRC information
• Example: "Data mining your website
• Customization approach - Web lists by subject:
• Mathematics (list at University of Alabama Rodgers Library)

Faceted Classification (Non-hierarchical)

• Can be a part of a comprehensive system, e.g. Colon Classification
• Ranganathan, PMEST and the Colon Classification 
• Fungal diseases in the rice crops of Madras, 1950-1959: J381,4:433.441'N5
• Can be a part of a hierarchical system as a non-hierarchical specification of the aspects of a subject:
• DDC and LCC have tables for geographic and other facts
• NCSU Catalog
• Home Depot
• Often Used for web organization:
• Iokio CameraFinder

Social Classification

• User added metadata
• Shared resources (commonly used for collections of photos and URLs)
• Organized via third party collaborative websites.
• Also known as folksonomy, ethnoclassification and free-tagging.
• Tagging - the establishment of a relationship between an online resource and a user:
• No centralized vocabulary control
• However, intent is to match an individual with other individuals who not only have the same interest, but also share the same way to express the aboutness of that resource (let "pop" be pop and "soda" be soda)
• Contexts (know what is being organzied!) [ e.g. Photographs]
• Virtual Photographic "Shoeboxes"
• Collections of digital photographs stored at a third party website
• Users "Tag" their photos with descriptors and descriptors can be searched.
• Lack of vocabulary control; however, in this "social" context, netiher precision nor recall is important. 
• EXAMPLE: third party photo aggregator: Flickr

Know your Resource Collection!

• What type of resource?
• Pages or collection of pages?
• Webpages, articles, books or other?
• Are they structured consistently?
• Do they follow conventions of publishing?
• How are resources organized?
• Searchable surrogates?
• Authority control?
• Automatically indexed?
• Human (professional or civilian) indexer?
• Browsable directly?
• Both searchable and browseable?