Third-Party Tagging: Next Stop, Google Groups?
I wrote earlier about Google’s effort to enlist users to tag images, in part through a competitive game that offers an incentive system.
Notwithstanding the likelihood that participation in the mind-numbing guessing game will slow (I’m happy to report I just placed 53rd overall in my first round of play today, and can’t for the life of me think why I’d go back for another), it makes you think about all the stuff out there that still needs to be tagged… and how that’s going to happen.
You can see how this would have wider applications, even just within Google’s stable of properties.
Newsgroup postings on Google Groups, for example, aren’t tagged in that Web 2.0 fashion yet. Many useful posts don’t always come up in a search, or would at least be easier to find if users tagged them with a variety of related tags. The less onus on the user to come up with the perfect search query, the better… and when it comes to particularly useful newsgroup posts, the community can help. To reiterate, that model is not too much different from how the community (sometimes unwittingly) helped to create a pretty accurate third-party tagging and evaluating system by linking to “useful or related pages and sites” in the hyperlinked environment, often by including relevant anchor text or relevant nearby text. The usefulness of this has diminished somewhat due to “gaming” (search engine optimization tactics), but it’s still considered foundational to what makes the web tick.
The newer forms of third-party classification are, to a large extent, simply extending that age-old hyperlinks-with-relevant-anchor-text model. It was actually pretty clever for search engines, notably Google and Teoma, to glom onto this “world of third party opinion via linkage” as the dominant approach to “metadata,” replacing the old kinds of keyword metadata (which were and are, ironically, called keyword meta tags). The key differences between third-party hyperlinking behavior and first-party metatag decisions were (1) the “party” – third is more credible than first; (2) the ability to classify stuff even when the creators are too lazy to do so themselves… and to do so without having to rely on formal editors or gatekeepers, but rather the wider community. As we all know, that led to contrived activity (such as linking “campaigns”) that carries on to this day, but that’s another chapter.
The scope of this activity seems to rest heavily on an incentive system, but it needn’t. The particularly useful stuff can get tagged (doesn’t the notable content already always get that extra oomph from the user community, whether that means it gets linked to, bookmarked, or dugg?). The rest needn’t be. Everyone wins.
In 1999 Andrew co-founded Traffick.com, an acclaimed “guide to portals” which foresaw the rise of trends such as paid search and semantic analysis.