Here’s a view of upcoming events in London at this time.

Three closely related topics with a number of web-savvy taggers (if attendees of events like this can’t tag, who can?). And yet, zero tags in common.

It would be very compelling to subscribe to events in your area and under a certain topic, so they automatically appear in ICal, GCal, etc. But the topic would have to be general enough to provide more than one event a year, and chances are there won’t be many “microformats” gigs in your local area on a regular basis. i.e. what’s obviously required is a tag like “geek” or “web” that’s general enough to be useful for the use case of a calendar subscriber.

Perhaps reflecting the tendency for people to think in concrete terms, the taggers here opt for very precise, specific, terms. The conventional wisdom on tagging is that people will come up with their own conventions and systems should stay out of it, but isn’t that a rather defeatist stance on our ability to create remotely intelligent software in 2006? Rather than force users to go against their nature and think abstractly, maybe the system could go all fuzzy and do some useful cluster analysis. I’d rather subscribe to a slightly inaccurate “web-ish” cluster than 20 different feeds covering every concrete concept related to the web. Granted, it’s also more work to develop, but it’s something that could easily be encapsulated in a library.