I’ve Heard of Folksonomies. Now How do I Apply them to My Site?

First off, let me talk about a major issue that we’ve been having with folksonomies: the very definition of the word.

Thomas Vander Wal, who coined the term, defines folksonomy as “regular people’s categorization through tagging”. Peter Van Dijck explains folksonomy this way: “Folksonomies are taxonomies created by users who add tags to things”. Another early explorer of folksonomies, Adam Mathes, calls it “cooperative classification and communication through shared metadata”.

So in practice folksonomy includes the tagging of things and the resulting taxonomy that arises from it. Notice, though, that there are two distinct activities here. One activity is tagging, or adding metadata. The other activity is the creation of a taxonomy based on those tags.

Tagging, by itself, does not a folksonomy make. It is possible, as Clay Shirky has pointed out, to tag things without creating a folksonomy. Tagging is simply an explicit activity that people can do to add metadata to content. It is common. Information architects tag things. I tag things on my computer. Every web page consists of dozens of tags (albeit with little meaning). In general, creating metadata includes a lot of tagging. Tagging as an activity is neither unique nor special.

Because we can aggregate tags, however, we can build a taxonomy out of them. More specifically, we can build a taxonomy out of the patterns we see in how people use tags. It is this act of aggregation, and not the act of tagging, that give folksonomies their power. Without aggregation, tags are just tags, with no meaning beyond the local meaning that each user gives to their own set.

Although the two most talked-about folksonomies, del.icio.us and flickr, use tags, we don’t need them to create similar, powerful tools. All we need is something that we can aggregate.

So instead of asking “how can I create a tagging system for my site?”, we might ask: “how can I aggregate and leverage the behavior of people on my site?”. When asked in this way, we recognize many examples of this being done right now.

Take, for example, the most emailed articles functionality from the New York Times online. What they do is aggregate the activity of people emailing articles and then create a list out of them so others can see or even subscribe to via RSS. This benefits both the people using the site (they can find out what’s most popular) as well as the New York Times (they can find out what content is most effective). Though not based on the activity of tagging, and without building a site-wide taxonomy, this functionality leverages aggregate human behavior in much the same way that folksonomies do.

Another example is the personalization features of Amazon.com. These are full of useful aggregation techniques. Each time you view a product page, for instance, Amazon shows you products that other people (who either bought or viewed the same product) were interested in. The activity aggregated here, instead of tagging, is the viewing of a product page. It is explicit, can be aggregated, and easy to see patterns in (though probably very difficult to implement given the size of Amazon’s inventory).

The leading RSS readers do something similar, too. Bloglines, for example, has a great recommendations feature that lets you find out what feeds are related to each other by aggregating which feeds are often subscribed to by the same person. In my experience, their recommendations are very good.

So don’t let tags fool you. Tagging is the activity that allows del.icio.us and flickr to provide useful functionality, but it may not be what works on your site. For you it might be most-emailed articles or “people who viewed this also viewed that” type of thing.

The main thing is that the activity you choose to observe needs to be explicit, can be aggregated, and important to the needs of your site.