Folksonomies: A User-Driven Approach to Organizing Content

Thanks to Patric Conrad for the a Belorussian translation of this article.

Many of the design teams we talk to face the same major issue: how to organize the information on their sites. From creating navigation schemes to developing site hierarchies to refining checkout sequences, it’s highly important for design teams to organize information effectively for their users.

Information architects frequently must deal with the problem of managing more and more information. Rarely can they remove information from a site; in most cases, it’s add, add, add. Design teams must make room for this new content in some way, either by incorporating it into the current organizational scheme or by altering the information architecture to allow for it.

The Frustration with Taxonomies

One of the most common strategies for organizing content is to place it in a taxonomy. A taxonomy is a hierarchical tree structure such as those used in scientific classification schemes. For example, the taxonomy for organizing all living things has 7 taxons, or levels: kingdom, phylum, class, order, family, genus, and species. Since a taxonomy is hierarchical, each living thing (each species) is organized under a particular genus, which in turn is organized under a particular family, which in turn is organized under a particular order, and so on up the tree.

This classification of living things works well for scientists because everyone is dealing with the same information: the information about living things. Information architects, whose taxonomies are revealed in navigation schemes and site maps, don’t have this luxury, however. Every web site contains its own unique information, so there is no single classification scheme that works for everybody.

Although taxonomies are common, it can be difficult for design teams to implement them. For one thing, taxonomies are very expensive to create and maintain, often involving month-long projects by several members of the team. For sites with thousands (or even millions) of pages, this Herculean task is sometimes never complete. As a result, broken taxonomies can remain until the design team attempts a complete redesign. Second, taxonomies may fail to reflect the language of users if they are not fully tested with the target population. This results in a less effective site that leads to user failure, user frustration, or increased support costs.

Folksonomies: A New Approach

Folksonomies, a new user-driven approach to organizing information, may help alleviate some of the challenges of taxonomies. Sites with folksonomies include two basic capabilities: they let users add “tags” to information and they create navigational links out of those tags to help users find and organize that information later.

For example, on the photo-sharing site Flickr, a user can tag each of their humorous pictures from a vacation with the words “hawaii” and “funny”. Some other pictures they can tag “hawaii”, but not “funny”. Others still they can tag “funny” but not “hawaii”. Later, when they want to view their pictures from Hawaii, they can visit a page consisting of all the tags they’ve used, and simply click on “hawaii”. If they want to view all their funny pictures instead (including, but not limited to, ones from Hawaii), they would click on “funny”. If they want to view only their funny pictures from Hawaii, they would simply click the checkbox next to both “hawaii” and “funny”.

Folksonomies Address Taxonomy Difficulties

In this way Flickr’s folksonomy addresses two of the most difficult problems with taxonomies. The information within folksonomies is organized and maintained by users, so very little work has to be done by designers after initially setting up the tagging system. This could be a boon to information architects who now spend too much of their time regularly re-organizing their taxonomy.

One of the most promising features of folksonomies is that there is no disconnect between the user’s words and the words on the site: the users words are the words on the site! Not only are users able to organize their stuff according to their own rules, but the information architects of the site can learn interesting things that a taxonomy may not have illuminated. For example, on Flickr one of the most popular tags is the word “cameraphone”, used to tag pictures that people have taken with the products. This is evidence that users are very comfortable with that term, despite it being only a few years old. Thus, information architects can include this term in appropriate places on the site knowing that users will be familiar with it.

Applying Folksonomies in Other Areas is Untested

Folksonomies aren’t a cure-all, however. Because we only have a handful of working folksonomies (others include Del.icio.us, 43things, and CiteULike), there are many questions left to answer about them. Perhaps the most interesting question is how to apply them to an existing site. Since Flickr is a specialized photo-sharing application, it’s not immediately clear how a corporate site, financial site, or shopping site might make use of the same organizational scheme. On the other hand, we’ve talked with several organizations whose information architects are at their wits end with taxonomies, and so testing the folksonomic waters might be a welcome research opportunity.

We’re Optimistic about Folksonomies

At this point, folksonomies are more of an interesting technology than a tried-and-true design tool. However, with their ability to let users do most of the organizational work of the information on a web site, they may yet prove to be a valuable, time-saving way for information architects to keep a handle on the addition of information into an already-burdened architecture. We’re anxious to see where they go.