"Folksonomy" at InfoWorld

I don't have enough time to do a regular blog, so I figure I'll start an occasional series here.

InfoWorld made a little splash in the blogosphere with word that it was giving up taxonomy in favor of tagging. Designer Matt McAlister explains in his blog how they are "excited about the possibilities for the site now that we have these tags", which are powered by del.icio.us.

First off, he says they are going to be combining "structured" tags applied in a "normalized" way with "free-form" tags. By "structured", he means that they'll have a controlled vocabulary where it's important, for instance so that ads can be sold against certain content. Although I'm not sure where it wouldn't be important.

He's excited that they're going to be able to find related content by looking at content that shares more than one tag. Which means they've re-invented post-coordinate search.

What most annoys me is that he says that "The downside is that we're probably going to phase out or at least simplify the robust taxonomy that we spent so much time and energy building and refining over the years." That is a downside, because there's no point to it. A more flexible view of their taxonomy, making it more hospitable to new topics, applying more than one topic to a story, and moving towards faceted classification would provide them with the benefits they seek without throwing out all their previous work.

Now, a peek at their tagging in practice. Look at del.icio.us/infoworld. (Disclaimer: they've just started doing this, so perhaps there's a learning curve, which I'm not giving them credit for). There's lots of use fragmentation of content in their tags: app_server and application_server, bigfix and bigfix_patch_manager, rim and research_in_motion. There's misspelling: delyaed (for delayed). There's inconsistency: 32_bit and 64-bit. There's splitting names among two tags: Carly and Fiorina (that might work) or Mark and Hurd (that won't).

The result is chaos, which a controlled vocabulary or taxonomy would prevent. Other than trendiness, there doesn't seem to be much value here in abandoning established practices. Jon Udell needs to walk down the hall and talk some sense into his colleagues.

Syndicate content