Folksonomy: A Game of High-tech (and High-stakes) Tag

Page 1 of 3

Should a robot dictate the terms of your search? In an age when whole lives are lived online—via blogs, picture albums, dating, shopping lists—digital content users are not only creating their content, they're building their own infrastructure for making it easier to find.

The term folksonomy was coined in 2005 when information architect Thomas Vander Wal mashed up the words taxonomy and folk to name the growing phenomenon of users generating metadata by tagging pieces of digital information with their own searchable keywords. The process is simple enough: Users assign a name, or tag, to any image, article, blog, bookmark, or URL. Later, when they want to recall this content, they can search for its tag and find exactly what they're looking for. Applications, sites, and new uses for tags have been spreading like wildfire, so much so that major Internet commerce companies are starting to invest serious attention (and funds) in making these home-grown taxonomies part of their systems.

If folksonomy sounds like anarchy on the World Wide Web, with people bending rules to their individual needs and tastes, it isn't. In fact, tagging is at the core of some of the most vibrant and cohesive online communities. 

Not everyone is ready to leave behind the structured comfort of a controlled taxonomy and jump on the self-tagging bandwagon just yet, however. For the same reason librarians still rely heavily on the Dewey decimal system, information architects argue that the Internet will always need a clear taxonomical structure to make digital surfing manageable. 

Many cybersearchers are pushing against the limitations of a traditional taxonomy. Whether they succeed in overthrowing the controlled vocabulary hierarchy and setting up a free-for-all folksonomy in its place remains to be seen, but experts agree—now that "every man" has power over the language of classification, searching will never be the same. 

Search Vocabulary in the Vernacular 
A vast amount of digital data goes online every day. When people tag that data, they aren't just creating more data—they're creating metadata, which is used by search engines to interpret the content to which it's attached. Major search engines create metadata based on predefined categories for every page that they index, which their individual searchbots can read. The problem is that content is being created faster than bots can crawl, extract keywords, and fit them into the engine's hierarchy. Not to mention the fact that a lot of content is made up of pictures, blog entries, personal Web sites, and other personal data that can't be easily crawled by searchbots, which only understand text. 

The major search engines cover a lot of ground, but simple searches yield far too many results to recall something specific. Using tags, the content that most users tagged, say, with the term bears will rise to the top of the list, so searchers can at least be sure that some human was thinking bears when they tagged that content. According to VP of business development Chris Fralic, tags are based on the user's personal search experience, "the first few words that come to mind when you're in a particular frame of mind," which "tends to be how you'll want to remember or discover the same thing or similar things in the future."

In the way that Google's metadata is designed to be readable by its searchbots, tags in folksonomies are intended to be read by humans. Users have control over which labels are applied to what, though most sites do suggest a kind of structure by prominently displaying the most popular tags and showing a list of other users' tags for any piece of content or bookmark. 

Broad Side
Folksonomies typically fall into one of two categories, as outlined by Vander Wal. The first, broad folksonomies, are created when third-party users assign various tags to the same content, essentially creating metadata for their bookmarks. Sites like and then aggregate this metadata and make it searchable. Users of these sites can see what other tags have been created for certain content and use this to get a broad idea of associated terms and tags. They're often referred to as social classifications, since seeing what other users are thinking about is as much a part of the site as finding what you need. The richness and participation of the communities around these folksonomies are, in large part, what fuel interest in them—it's like "eavesdropping on someone else's thought pattern," according to Hadley Reynolds, senior analyst and head of research for Perot Systems' Delphi Group.

Collaboration through collective tagging gives members of these communities a chance to build their own search systems from the ground up, based on their own vocabularies, interests, and ideas. Folksonomy sites use simple popularity (number of tags) to rank articles on their homepages, so users can easily sample what the rest of the community has been tagging. The vox populi rules on these sites; instead of exchanging words, members exchange their ideas by bringing either content or bookmarks into the community and promoting them., also known as a social bookmarking site, is one of the fastest growing folksonomy communities, with nearly 300,000 members and counting. When a user wants to tag a page, all he or she has to do is select it, choose a tag for it, and the bookmark is automatically stored under its assigned tags in the user's personal folder. The tag also becomes a part of the community at large, and members can search for tags within their individual folders as well as the entire community. 

"The social side of lets users discover what's important to other users," says Fralic. Subscribing to RSS feeds around certain tags and things like, the site's list of most-tagged items, provides the community a distinct, unique way to find new sites users are interested in. "You could think of it as starting with self-interest," he says, "but then through sharing bookmarks and tags with others it becomes ‘enlightened self-interest.'" 

Also helping users collaborate over tags (or bookmarklets) is SurfWax's Nextaris. Nextaris users can drop their tags into a folder with nearly 100MB of storage space that a predetermined group of people can access; additionally, users can browse other folders to see what gems have been found by people with similar research goals. "A natural kind of synthesis takes place," says SurfWax CEO Tom Holt of Nextaris' collaborative community. "There will always be a need for a hierarchical system, but in terms of folksonomy . . . the contribution of tags doesn't necessarily need a moderator to thrive."     

Page 1 of 3