May 24, 2005

Re: Remove Forebrain and Serve: Tag Clouds II

This morning I stumbled onto You're It! a blog on tagging via a co-worker, and started perusing the content within. Much interesting content on there, with some of the usual cast (Clay Shirky, David Weinberger, Peter Merholz are top of mind for me), and I happened across Jeffrey Zeldman Presents The Daily Report - Remove Forebrain and Serve: Tag Clouds II. There is quite a lively discussion going on there around the issues of formal taxonomies vs. folksonomies and issue of the mob vs. the professional, etc.

While I posted some comments in Zeldman's separate blogspot discussion area I'm repeating them here for my own permanent record, and to encourage discussion to branch out within our own context here, which tends to be more enterprise-focused than individually focused.

The comments I'd posted:

Jeffrey and all - Great discussion, seems to be generating far more depth than other circles stuck on the folksonomies wave.

I'll just throw a few cents in here, maybe a buck three fifty:

  1. Randomization can do wonders to avoid the pure herd mentality problem. Let's learn from other systems, like expertise location - where you don't want the TOP expert being the only resource used, or you will drive him/her insane and cease all other 'productive work.' Mix it up, and distribute the load a bit. Seeing popular terms and trends is a very interesting artifact of these systems, but not the main benefit, IMHO. This 'tag cloud' visualization business is also one heck of an eye sore from a usability standpoint. Lightweight certainly, but yeesh...
  2. The way that and others work allow (and encourage) people to have multiple tags, rather than the standard controlled vocabulary approach (in an enterprise deployment, not the web at large) which is typically (not always mind you) a SINGLE tag. With multiple tags comes the ability to cross-slice within an individual's collection or across collections. Personally, I wish there was an easy way to create synonym rings or a thesaurus in to squish the folksonomies a bit. Unlike Clay Shirky, I'm not convinced that film, movie, cinema people are from different planets and never intersect.
  3. When I talk about folksonomies and personal vocabs in our Proving Ground on Information Architecture and Taxonomy event, or our consulting work, 90-95% of the people I'm working with look at me like I have two heads, perhaps three (including my own colleagues - which I'm used to). They don't have the slightest clue what I'm talking about, and even after showing it and explaining it, they don't quite get it. Let's not forget that organizations/enterprises move just a bit slower than the cutting edge blogosphere - with any luck that delay lets organizations skip out on this messy beta phase and straight to the next major stepping stone.
  4. Anyone who believes that controlled vocabularies OR folksonomies OR search OR collaborative ranking OR yadda yadda are THE ONE SOLUTION in and of themselves should really take a break from drinking the kool-aid, and get some fresh air. The solution depends on the problem at hand, which depends on context, content, and the community being served. Folksonomies, while ostensibly being about sharing, seem to be about tracking your own important things FIRST, and sharing secondarily, and both of those aspects are still not quite fully baked.
  5. The idea that Clay Shirky had mentioned on using these services to keep found things found is great, except that and most of these systems (outside of Furl, or is it Spurl?) don't keep copies of the content frozen in time, it's just a pointer. Outside of blogs, most systems don't have permalinks, and many sites are pulling dynamic content of one sort or another. Perhaps it's time for Google to leverage their caching mechanisms, along with social bookmarking, their new portal offering (back to the future anyone?), and their traditional search/pagerank capability into the next generation of findability/searchability, fully of Googly goodness?
  6. The inability to search on actual content within et al is a right pain... I like being able to demonstrate the pros and cons of pure tagging vs. pure search and the intersection, so showing vs. Cocoalicious helps to easily illustrate that. Until people can see directly what the good and bad are of these capabilities, it just doesn't sink in. To my earlier point, no single one of these options is likely to be THE solution.
  7. The whole business of all of these social bookmarking systems using their own secret tagging language of spaces, not spaces, commas, CamelCase, etc. - 'normal' people don't want to have to deal with this craziness. This is a real problem for even broader adoption. I'm willing to do it, but then again, I'm a sucker for experimentation.

I'm curious - anyone using these ideas within the walls of their own systems/organizations, or is it just out here in the wild west that folksonomies/social bookmarking is happening?


Posted by Dan Keldsen on May 24, 2005 | Permalink | Comments (1) | TrackBack