SXSWi Day Two: Tagginess

The highlight of SXSW Interactive Day Two, apart from a confounding mock cosplay battle outside the front entrance to the Convention Center (promotion for a documentary screening at SXSW Film), was the morning's Tagging 2.0 discussion. With all this buzz surrounding the so-called Web 2.0 -- which, we're loath to admit, we still don't actually understand -- moderator Don Turnbull thought it prudent to use a similar catch-phrase, as both emerging technologies (or concepts) seek to expand the user-driven online experience. Tags, which we've recently implemented in our SXSW Music coverage, are a form of "metadata." In the construct that most of the blogging community utilizes, they're useful for linking together various blogs and information sources that share a similar theme, concept, or focus. A food-related entry on Austinist focused on spanakopita, for example, could be tagged "greekfood," "lovefetacheese," "appetizer," or itself, and in such a fashion we might merge these tags with the aggregates on Technorati, Shadows, or any one of the burgeoning websites catering to such information trafficking. What results is something akin to a Google search, only whereas Google trolls the internet with automated "spiders" and finds what it deems relevant (through complicated, proprietary algorithms that we're not sure are all that sophisticated, given the propensity for some people to stumble upon our humble blog after googling "free male porn" or "real world nudie pics!!!" -- seriously), tags are created by the writers and/or readers of a particular site and thusly that much more appropriate to the subject being discussed.
But back to the panel discussion: five different speakers each discussed their view of what the future of tagging might entail, sparked by Turnbull's opening remark asking what tags were useful for, and how they might be improved.
Adina Levin, a Vice President over at Socialtext, advocated the value of tags for their intrinsic social nature, despite the somewhat clumsy semantics (plural/singular tags?). But complicating the creation of tags with a set of rigorous requirements, she warned, would do more damage than good. Users, after all, want the simplest interface that can accomplish a minimal set of tasks. In our opinion, it seems like the burden of sifting through obfuscated tag branches should rest on the shoulders of the aggregation systems out there.
Our favorite of the talks happed to be from Shadows.com's "Tech Evangelist" Prentiss Riddle, who also runs a fairly prominent personal blog called aprendiz de todo, maestro de nada. Opting to go against the grain, he titled his presentation "The Six Dirty Secrets of Tagging," which were (thanks to the timely recap on his blog):
1. It’s the content, stupid. Tags are a means to an end, not an end themselves.
2. Ordinary users don’t understand tags. Present a normal person with a tag box and they’ll ignore it or enter an English sentence.
3. It’s the UX, stupid. When tagging systems do work it’s because a great deal of care went into the end-to-end user experience. Flickr is a good reference point here.
4. Tags don’t play well with others. Tagging systems are plagued with interoperability problems, like differing standards for delimiters and different social norms.
5. Rich applications require rich metadata. Or, where’s my flying car?
6. Nobody likes real tags. What they want is tagginess!
The rest of his presentation is posted online.
Thomas Vander Wal, Principal Consultant of InfoCloud Solutions, came up with the term "folksonomy." Wikipedia has it defined as a "collaboratively generated, open-ended labeling system that enables Internet users to categorize content such as Web pages, online photographs, and Web links." With dozens of sites now implementing so-called "social bookmarking," and massive e-commerce entities jumping on the tagging bandwagon, it's clear that tags -- which are less an actual technology as they are an abstraction of an organizational idea -- are being embraced by the wider public. Vander Wal later introduced the notion of a multifaceted folksonomy triad, through which the community and tagger are connected.
Lastly, Rashmi Sinha of Uzanto confidently whisked through a presentation more focused on the sociological implications of an internet community connected through tags. Recalling the philosophy of Joshua Schachter, creator of del.icio.us, who basically said that one should simply commit to paper (or fiddly computer bits) the first things that one thinks of when tagging a particular item, Sinha stressed that tags were the web's best attempt at appealing to the greatest majority: inherent to its communal nature were four key advantages (cognitive diversity, independence, decentralization, easy aggregation).
Ultimately, where does one go from here? We're not sure, though the presenters did bring up several of the disadvantages of tags, as they exist now, as a means of further exploration.


