In the context of social software applications like blogs, there are two kinds of tags: categories/in line keywords and collaborative tags. Josh Ledgard seems to be talking about the former, owner-controlled tagging, in his recent post. The latter, reader-contributed tagging (which I refer to as "collaborative tagging" in this post), was one of the main topics of discussion at the Microsoft Research Social Computing Symposium earlier this month (event tags: SCS2005 (Flickr) and SCS2005 (Technorati). Collaborative tagging empowers readers to assign keywords to other peoples' content, like this blog post, and thereby associate it with other pieces of content, the creator/owners of which do not necessarily even know about each other or consider their content to be related. In a presentation about collaborative tagging at the symposium, David Weinberger said,

"Knowledge can find its natural shape now that it has been freed from the physical [tree].”

I think that what he's really talking about is control. People who create content generally "own" and organize their information (like me with this blog) in ways that knowingly or unknowingly constrain its significance in context and limit its discoverability.

For the purpose of this post, I define tag as a label and associate it with the keyword "Social Computing" by way of adding this post to a category of like name. Implicitly, I am asserting that the word tag belongs to my personal "Social Computing" bucket. But does it? Taxonomists would certainly disagree. Josh might disagree. There's nothing social about tagging a Web site by adding it to a category such as a favorites folder. Readers who are interested in what I have to say about Social Computing can more readily find this post by subscribing to or browsing through my "Social Computing" category. Readers who are interested in semantics, knowledge management (KM), taxonomies, and many other things, will probably not discover it so readily. But if I allow my readers, some of whom are interested in BOTH social computing AND semantics/KM/taxonomies, to assign the word "tag" to this post, other potential readers who are not interested in social computing will have a much better chance of happening upon and deriving value from my blog post than they do now. As a consequence of looking at my referrers log, I might even be enlivened to new ideas, projects, people who share my interests, and cross-currents of thought across multiple disciplines. This fluid, serendipitous dynamic liberates knowledge from the constraints of my limited perspective.

Knowledge is unformed, non-hierarchical, and organic. When we try, as is our instinct and training, to force knowledge into a well-formed, chronological, hierarchical, and conclusive shape, we constrain it. Honestly, I don't know who will be interested in this blog post or why. So why should I constrain its potential to spark a conversation or connect me to other people who share my interests? By empowering readers to associate ideamarkers with the knowledge I present online, my limited knowledge has the potential to grow into something that is infinitely more powerful, meaningful, universal, and discoverable than I alone can make it.

Social software geeks refer to a collection of collaborative tags as a "folksonomy." In German, that would be a Volksonomy. I hope to one day have a chance to take a Volksonomy for a test drive on What tag would you assign to this post if you could do so? No expletives or pejoratives please. ;-)

Note: Für meine deutschen Leser, kein ist es nicht Blog *Tag*. :-) Nein, es "Tagging" Tag, der schon besser ist.