• by Tom Hespos , May 2, 2006

I don't think anyone would disagree that we're in the middle of a content explosion. A combination of factors, including the ease of use of content publishing tools, increased penetration of broadband Internet access and the level playing field provided by the Internet has led to a huge deluge of independently produced content. Along with that content comes its own method of categorization: social tagging.

Social tagging represents an efficient and accurate way to categorize content as it is created, or as it is linked. According to Technorati, nearly half of all blog posts are associated with one or more author-generated tags. This categorization forms the basis for a system that can determine content's relevance to a given subject. In other words, this could be the next wave in search, but it's also much more than that.

Let's say I buy a new mobile phone, and I decide to blog about the phone, describing its features and my likes and dislikes. I put together my post, and just as I'm ready to post it to my blog (and thus, the world), I assign any number of tags to it so that people searching for content related to those tags will be able to find it more easily. Let's say I assign "Motorola," "camera phones," "mobile devices" and a number of other relevant tags to the post. Now, I've categorized my post and declared its relevance to a number of different terms.

That's not the end of it. Someone else may see my post and decide that my comments about the phone's well-written user manual represent something that the world of technical writers ought to see. He links to my post about the phone, but categorizes it under "good technical writing."

As you can see, social tagging represents a scalable way to attack the Herculean task of categorizing the content distributed on the Internet. Looking beyond that basic function, though, you can also see how social tagging democratically determines relevance to a given subject. It makes sense; now that we have a distributed network on which to share content (the Internet), we need a corresponding distributed system for categorizing that content.

One of the comparative advantages that social tagging has over automated indexing like search engine databases is that it relies less on surrogates. Search engine relevance algorithms depend on how many times a particular keyword is mentioned in an article, or on how many sites link to a given piece of content. But what about the example I laid out earlier, where I never mention the term "technical writing" in my review of a particular device, but someone else thinks the post is relevant to that term? Search engines currently can't do that well. With scale, social tagging has the potential to do it better.

What's terrific is that that so much of the content being published on the Internet is tagged, and that which hasn't been tagged yet can be categorized according to how folks link to it.

Before you dismiss this and think that it's yet another quirk of the blogosphere that can be easily ignored, know that social tagging is bigger than the blogosphere and even the citizen publishing movement. Since anything online can be categorized in any number of different ways, this has relevance across the entire Internet, whether it's classifying photos on Flickr or posts within your favorite RSS feed. I'd encourage you to check out the how-to page on sites like del.icio.us and read about social tagging on Wikipedia for more information.