I'm puzzled by people who file things in del.icio.us under interesting. If it wasn't interesting, they wouldn't bookmark it in the first place, right ?
Anyway. I understand that the purpose of the free tagging devices like del.icio.us (the tags being free, not necessarily the service) is to try to spontaneously come up with shared and useful meanings for tags, and that it's supposed to work if people just randomly sick in two or three keywords to each item they categorize. -- However, my brain is not wired this way. I need to have a consistent set of tags that I can manage and organize as a whole. Am I doomed ?Not really.Am I doomed?
The problem is that classification is a very tough problem to solve. You could spend lots of money and lots of time coming up with the one true classification system that will always be right (except when it disagrees with every other classification system that other people have spent lots of time and money developing).
del.icio.us is a Perlish response: the solution is messy because the problem space is messy. Instead of solving the problem on the scale of 1, 2, or 3 classification systems, it solves the problem on the scale of a few thousand classification systems. Sure there will be noise, but over time, the chaos reaches a local maximum, and you begin to see the patterns that are meaningful to you. After a while, you can identify patterns in the noise that help you filter them out. (i.e., ignore bookmarks from anyone who uses the tag 'interesting'.)
The only problem with del.icio.us folksonomy at the moment is that Joshua needs to spend some more time on filtering, bundling, merging and profiling links and tags. But these are the kinds of questions that only arise when you have a few million bookmarks to study. We're just now starting to see both the problems and solutions emerge.
It discusses the problem you're having: predefined classifications, AKA "controllled vocabularies", vs. the free tagging as on del.icio.us, aka "folksonomies". Like Ziggy said, people are only starting to grasp how it works, how it should work, how it should evolve.
The major problem with controlled vocabularies is, and I quote from that article:
Furthermore, users pollute controlled vocabularies, either because they misapply the words, or stretch them to uses the designers never imagined, or because the designers say “Oh, let’s throw in an ‘Other’ category, as a fail-safe” which then balloons so far out of control that most of what gets filed gets filed in the junk drawer. Usenet blew up in exactly this fashion, where the 7 top-level controlled categories were extended to include an 8th, the ‘alt.’ hierarchy, which exploded and came to dwarf the entire, sanctioned corpus of groups.
The major problem of folksonomies, OTOH, is the complete lack of structure, and synonymous tags. Merging those tags will most likely be one of the things we'll have to apply in the future — the major problem yet to tackle being partly overlapping tags, like "math" and "calculus", where the latter is a subpart of the former. In theory you can tackle that problem by attaching more than one tag (a major advantage a free tagging system, as opposed to how a book can only be in one place in the library), but it remains a problem when people forget about the more general tag in favour of the more specific one.
BTW one more interesting article is Tagging the Internet, an article about the promise of folksonomies.
