Why I love Twitter

I've been in the taxonomy business for a long time; in fact, I've been working on vocabulary tools, and "indexing" documents for online information systems for longer than I care to admit. So you'd think that I'd be a fan of the traditional controlled vocabulary, including a nice, balanced subject hierarchy, managed centrally. But I'm not. While I like building hierarchies, I've always found them to be too restrictive and not organic or natural enough, not always reflective of common parlance, and often not at all aware of user context. So that's where Twitter comes in.

While it's fun to read about what my friends are eating (and that happens a lot on Twitter) it's much more fun to see real-time tagging, including context and verbs, about real-time events, which is what I find on Twitter. For example, I "follow" alaska_avo on Twitter. I don't know who alaska_avo is, but I do know that I'll get real time information about the state of the Redoubt Volcano, every time alaska_avo posts. Their name provides context, and their tweet immediately tells me what I want to know. A typical tweet starts like this: "Redoubt: Redoubt Volcano has not erupted..." and almost always provides a URL where I can go to learn more. For my money, these are the most informative "tags" out there because they're context-driven, and because they're actionable. Picture a corporate "Twitter" site (I'm sure someone has already done this, so let me know.) It would essentially enhance tagging by including context, and it'd be in real-time. A fictional HR_Speaks, for example, might tweet "Benefits: Open enrollment starts today, see http:\\intranetHR_URL.com" with a link to the corporate HR site.

I also like to follow the hashtags on Twitter, although they make me laugh because they are so cryptic. In December, we had record snowfall here in Seattle, but we live in the land of the micro-climate, so the snow experience was different depending on where you live. Twitterers created a hashtag, #seatst, which provided real-time tweets about the local conditions. (You'd type #seatst into their Search bar to see all the tweets that referenced the hashtag, updated as new Tweets appeared.) In this case, the user names didn't help me with context, unless they included the posters' location, but the hashtag sure did. (Snowy_In_Redmond, for example.)

The irony here is that I expect we'll see requests for a consistent set of tags soon, to replace the cryptic hashtag names, but I'm going to wait and see how that develops. But picture a context-driven "taxonomy", a la Twitter, and think about how useful that would be...