The process of classification and taxonomy has, until recently, been the exclusive responsibility of the librarian operating in the familiar systems of Dewey Decimal and Library of Congress (LOC). Large databases, such as those for journals, have their own internally-controlled vocabulary. Classic taxonomy is characteristically static and often cryptic, leaving the effort of learning these systems to the user. It is incumbent on the patrons to familiarize themselves with the major classification schemes and database indices in order to navigate through to the desired information.
In response to such constrictions, tagging, folksonomies, and social software invert the process of creating metadata: instead of making the information fit a category, they create a tag to fit the information. The overarching idea in applying social bookmarking is that the user can tag items with words that are meaningful to them. This bottom-up method of organizing information is commonly called a folksonomy. Folksonomies are often adopted for web information as the scale of information is beyond any individual or organization to classify. By making the organization of information a collaborative process between librarians and users, the intent is to make navigating the countless information resources available a more intuitive process.
The Internet is evolving into a user-centred, user-designed universe, where all individuals can contribute for their own benefit as well as the benefit of others. The great advantage here is that users can tailor or even design the applications and the organizational structure to suit their own needs. There are fewer constraints in locating information, as individuals can tag items with words that are meaningful to them, instead of being restricted to the closest match they can find in a controlled vocabulary with which the average user is often not familiar.
One way in which users can participate is through the use of social bookmarking applications. Individuals have the ability to access these applications wherever there is an Internet connection, they may add new content and apply meaningful tags, and these tags become part of the folksonomy.
Tagging is simply the process of applying a metadata keyword to information, whether it is a journal article or a website, to succinctly describe it and aid in search and retrieval. Tagging represents a radical shift from traditional taxonomies, which are hierarchical and exclusive, to a non-hierarchical and non-exclusive flat system (Golder and Huberman, 2006). For instance, the works of Confucius would be organized under philosophy, which would in turn be subcategorized under arts. A tag for items on Confucius could be “Chinese philosophy” or “Confucianism”. Tagging allows users to organize information according to their own cognitive models to increase findability and serendipity (Fox, 2006). The disadvantage of such a system is that people can have unique conceptualizations of things given the nebulousness of linguistic and cognitive boundaries, allowing for idiosyncratic personal categories (Golder and Huberman, 2006).
The word “folksonomy” is a portmanteau of the words “folk” and “taxonomy.” Traditional taxonomies organize information through pre-defined categories chosen by an expert in classification and categorization. The terminology is designed to attain consistency through controlled vocabularies. A taxonomy is often hierarchical, with a rigid organizational scheme that is visible to the user. Most notably, a taxonomy-driven organization of information is designed for expert users who have experience with the structure and are familiar with the controlled vocabulary.
One can immediately see how such a system would not map onto the dynamic and exponentially-growing information on the Internet. The uncontrolled development of information does not lend itself to a controlled organizational scheme. Thomas Vander Wal recently noted in his blog that enterprise organizations “have been finding 30 percent to nearly 70 percent of the terms used in tagging are not in their taxonomy.” Evidently, a new method of organization is needed to tackle the wide world of the web.
This need was answered through folksonomies. The word was coined by Vander Wal in 2004 in reference to the “user-created bottom-up categorical structure development with an emergent thesaurus.” It involves regular users (folk) developing a method for the organization of information (taxonomy) through the use of tagging. Some examples of toolss where folksonomies are applied are del.icio.us, Furl, Technorati and Flickr (although there is some debate over whether Flickr is a good example of a folksonomy).
Rather than hierarchical, folksonomies have a flat structure: relationships between terms are not labeled as “broader” or “narrower,” but are equally weighted. Folksonomies are user-generated: rather than an indexer or author attaching metadata to information, the user attaches tags that are meaningful to them for their retrieving needs. Finally, folksonomies occur in a social environment. While a user may tag information for personal use, their metadata (and often their identity) is visible to others for reference, search and retrieval.
Critics of folksonomies point out that user tagging creates meta noise through the inconsistencies in tagging: incorrect spellings, singular and plural tags, and case insensitivity. Searching, therefore, becomes more difficult, as a user must search the tags “folksonomy” and “folksonomies” to retrieve all tagged information. Retrieval is further complicated by the lack of synonym control; for example, mac, macintosh and apple are used interchangeably. Finally, the lack of hierarchy prevents parent-child relationships from being expressed. For example, a user can search the tag “social software” and not return items tagged with social bookmarking, wikis, blogs, RSS, or folksonomies.
However, the benefits to a user-driven, bottom-up taxonomy are undeniable. While many taxonomies require constant vigilance in selecting terms that will be relevant to the user, a folksonomy is created solely from user-selected, user-relevant terms. Folksonomies are inclusive in terminology, allowing anyone to add a personally-relevant tag to an item, regardless of its applicability or usefulness to the general population. As more users tag information popular terms and relationships are identified, providing effective search and retrieval.
Social bookmarking is the practice of saving bookmarks to a public Web site and “tagging” them with keywords. Bookmarking, on the other hand, is the practice of saving the address of a Web site you wish to visit in the future on your computer. In order to gain a full understanding of the topic, it may be useful to review earlier incarnations of bookmarking.
The first graphical browsers to become available all contained some means for individual users to keep track of their favourite sites, using a hotlist (in Mosaic), bookmarks (in Netscape) or favorites (in Internet Explorer). However, as the web has expanded, these simple hierarchical tools are no longer able to handle the complex variety of subjects available. As well, these tools for facilitating one’s web experience are only accessible from a particular computer, when using a specific browser. As search engines became more developed, it was faster to “Google it” than to find where the bookmark had been filed.
As more individuals became involved in interactive tools, these tools have become more useful for participants. Ebay and Wikipedia are obvious examples. The more items posted for sale, or the more entries defined, the more individuals will be tempted to shop or browse using these instruments. User participation breeds the success of the tool.
As users began participating more and more in contributing content, tagging it, and telling others about what they found on the web, the tools of social bookmarking have also developed, allowing users to create their own link management without the constraints common to cataloguing. As well, the ability to access these bookmarks from any computer, using any browser, makes social bookmarking a far more powerful tool than the original hotlists, bookmarks or favorites. Finally, the social aspect cannot be ignored. Although the system lacks privacy, it allows for greater exchange of information, as the links that are of interest to one person are likely to appeal to others as well.
Social bookmarking tools have developed over time, such that there are now tools to fit many different needs. Tools may use solely the free-form language of social tagging, and it is possible for the tools to also incorporate, and even automatically acquire, the controlled vocabulary of subject headings or the metadata of the articles themselves. It is easy to understand how this second option would be of great interest to academics and librarians alike.
While many of us are already familiar with some of the social tagging tools used primarily for social or recreational purposes,(e.g., del.icio.us, Flickr), the academic-focused tools are likely to be of greater interest in our future careers. The advantages of CiteULike and Connotea cannot be overstated. They truly do offer the best of both worlds, in being able to classify, and therefore search for, information in both a top-down (structured) and a bottom-up (unstructured) fashion. Originally, CiteULike only allowed users to bookmark articles located on specific websites, but it has since expanded its functionality to allow all articles into their system. As well, it allows users to import references from their reference software to their social bookmarking account. Connotea offers a similar product, with comparable features, and a more appealing interface, in our opinion. These two tools have great potential for any individuals who wish to conduct research.
The process of social bookmarking and the creation of folksonomies allows for the organization of information to become a two-way street: users the ability to apply their own conceptualizations on top of already existing and static vocabularies and taxonomies. Though still in its infancy in the library environment, its applications can be seen throughout the web from del.icio.us and Flickr to CiteULike and Connotea. There is potential to adopt this into other arenas within the library, such as OPACs and databases. Perhaps the integration of user-driven grassroots system of tagging will make the library’s contents easier to navigate and retrieve by the average user.
7 Thing You Should Know About Social Tagging from EduCause
Even Tastier del.icio.us by Philip J. Hollenback
Folksonomies: Power to the People by Emanuele Quintarelli
Folksonomies - Cooperative Classification and Communication Through Shared Metadata by Adam Mathes
Folksonomies: A User-Driven Approach to Organizing Content by Joshua Porter
Folksonomy article on Wikipedia
Off the Top, Thomas VanderWal’s Blog
Social Bookmarking Tools (I): A General Review by Hammond et al.
Social bookmarking article on Wikipedia
The Architecture of Participation by Tim O'Reilly
The Hive Mind: Folksonomies and User-based Tagging by Ellyssa Kroski
The Hype and the Hullabaloo of Web 2.0 by Ellyssa Kroski
Golder, Scott A., and Huberman, Bernardo A. “Usage patterns of collaborative tagging systems.” Journal of Information Science. Vol. 32, no.2 (2006): 198-206
Fox, Robert. “Cataloguing for the masses.” OCLC Systems & Services. Vol. 22, no.3 (2006): 166-172
By Christy, Jess and Neil